Anthropic Shuts Down Fable 5 and Mythos 5 AI Models Following US Government Directive

Anthropic shuts down Fable, Mythos models following Trump admin. directive

Anthropic completely shut off access to its Mythos 5 and Fable 5 models Friday night, just days after they were launched. The move comes after Anthropic's receipt of a US Commerce Department directive Friday evening, subjecting the new models to export controls restricting their use anywhere outside the United States. In a message posted Friday night, Anthropic said the only way for it to ensure compliance with that government order in the immediate term "is that we must abruptly disable Fable 5 and Mythos 5 for all our customers." Access to other Anthropic models is not affected. An Axios report cited an administration official saying that the administration is concerned by reports of a jailbreak that reportedly gets around broad classifier-based safeguards meant to block Fable 5 prompts regarding cybersecurity, chemistry, and biology. The administration reportedly requested a pause in the release of these models to gain time for the "national security apparatus" to be "hardened" against this kind of threat. That hardening could be complete "in the next few weeks," Axios' source suggested. In its Friday night announcement post, Anthropic said the government has only provided it with "verbal evidence of a potential narrow, non-universal jailbreak" that involves getting Fable 5 to review a specific codebase for software flaws. The company says it has only seen evidence of this kind of jailbreak being used to find "minor" and "relatively simple" software vulnerabilities, and that other publicly available models like GPT-5.5 has similar capabilities on this score. "We are complying with the government's legal directive and are removing access to Fable 5 and Mythos 5 for all users," Anthropic writes. "However, we disagree that the finding of a narrow potential jailbreak should be cause for recalling a commercial model deployed to hundreds of millions of people. If this standard was applied across the industry, we believe it would essentially halt all new model deployments for all frontier model providers." Earlier this month, President Trump signed an executive order urging AI model makers to submit to voluntary government security testing. That order came after an initial signing ceremony planned for last month was abruptly postponed amid reported concerns of disagreements about it within the administration. Anthropic apologized to customers for a "disruption" that it said is the result of a "misunderstanding," and said it will release more details about the situation in the next 24 hours.

Researchers caught in the crossfire as companies and government grapple over AI safety

For years, the firms pushing hardest to build ever more powerful artificial intelligence (AI) systems have also been among the loudest in warning about their potential dangers. Those competing impulses collided in spectacular fashion last week around Anthropic's Fable 5, one of the most capable models yet. Though many researchers had eagerly awaited its release on 9 June, the rollout quickly soured as they ran into sweeping restrictions on how the model could be used. But even those safeguards didn't satisfy the White House -- and just 3 days later it pulled the plug completely. For many scientists, the lesson was hard to miss: They're becoming increasingly dependent on AI research tools they do not control. That access instead rests in the hands of a small number of firms and policymakers who are still figuring out, in real time, how these powerful systems should be governed -- and what kind of research tasks they are allowed to do. This episode is "just the tip of the iceberg" in the struggle between regulating AI and giving free rein to research, says Rehaan Ahmad, co-founder of the open science research platform alphaXiv. "A reckoning here is yet to come." In April, Anthropic made waves with its announcement that it had made an AI model too dangerous to be released to the public. Mythos 5, the company claimed, was powerful enough to identify vulnerabilities in every major operating system and web browser in use today. A version of the model was only given to a select group of cyberdefenders and critical software infrastructure providers. So, researchers paid close attention when Anthropic announced plans to release a safer version called Fable 5, straitjacketed with extensive safeguards but also promoted as a major advance for scientific work. Anthropic had, for example, "heavily advertised Fable's biology capabilities," says biologist Sierra Bedwell, a Ph.D. student at the University of Illinois Urbana-Champaign. "The launch video was full of biological symbolism, and they had a post about how it hit all these biology benchmarks." Almost immediately, however, Bedwell and other biologists found they were effectively locked out. Fable 5 refused to answer even basic questions about biology, users reported on social media. Even a message comprising just two emojis, a germ and a DNA double helix, would trigger the model's safeguards. Some suspected the AI was using prior conversation history to identify users with biology backgrounds and apply stricter restrictions. Both AI firms and researchers worry AI might also be used for problematic purposes, such as building bioterror weapons. But Bedwell believes Anthropic's restrictions reflected too narrow a view of the risks and benefits of basic biology. She thinks the company could have come up with more practical limitations if it had worked with biologists in the months between the release of Mythos and Fable 5. Anthropic could have put "biologists on a team to actually talk about some of what legitimate risk questions would look like, rather than just saying 'Nope, we're cutting it off entirely,'" Bedwell says. "I don't think that was a good decision on their part." The AI research community was in an even bigger uproar. For cybersecurity, biology, and chemistry questions, Fable 5 at least made its restrictions transparent and told users they would be rerouted to a less powerful model. But the model's internal specifications revealed that if a query involved any sort of frontier AI research, it would silently degrade its capabilities in ways invisible to the user: essentially, secretly becoming dumber. The feature was likely intended to prevent competitors from using Anthropic's own systems to advance rival AI models. But for other researchers, it promised needless frustration and effectively put limits on what research someone could or couldn't do, critics argued. "If a model silently modifies or weakens its own answers while still pretending to help, researchers lose the ability to know whether a failed result came from their own idea, their implementation, or an invisible intervention by the model provider," Ahmad and alphaXiv co-founder Raj Palleti wrote in a statement on X. "Independent researchers, academic groups, startups, and open-source builders who rely on public tools to compete, innovate, and pioneer AI for everyone else" were likely to suffer the most, they added. The widespread backlash prompted Anthropic to quickly backpedal. It updated Fable 5's policy so refusals to answer AI queries would be visible rather than hidden. But the reprieve was short-lived. On 12 June, citing national security concerns, President Donald Trump's administration issued an export control order barring foreign nationals -- including Anthropic's own employees -- from using Fable and Mythos. To comply, Anthropic pulled both models offline. The move caught researchers off-guard. "It's not surprising that people are locking down models more, but I really didn't see the government [lockdown] coming," says Graham Neubig, a computer scientist at Carnegie Mellon University. For Sayash Kapoor, a computer scientist at Princeton University, the saga exposes a critical problem. As so-called frontier models become more powerful, independent researchers may know less and less about what those systems can actually do -- regardless of whether access is being limited by the companies developing them or by government intervention. "The key thing that these incidents bring to light is that there is a desperate and early need for an ecosystem of third-party evaluators," he says, so executives and policymakers aren't the only ones who fully understand a model's capabilities. Such independent evaluations are also important because frontier companies and governments are not necessarily asking the same questions as academics, Neubig adds. Some are optimistic that, with collaboration, scientists, companies, and government regulators can find ways to safely make the most powerful commercial AI models available to researchers. But others say the Fable 5 controversy also suggests researchers should be doubling down on developing alternatives. That idea has catalyzed calls to supercharge the building of open-source models, whose internal structures are publicly available and can be customized by individual scientists. Such models are generally thought to trail the leading commercial systems by 6 months to a year. But supporters argue they provide something increasingly valuable: a stable foundation for research that won't disappear at a moment's notice. Until then, scientists will be at the whim of AI companies who have the power to say, "'Oh, I don't want you working on this, I don't want you working on that,'" Ahmad says. "Basically, the only way that isn't a problem is if open-source research can at least put up some type of fight."

Anthropic's safety warnings may have just backfired -- the government has pulled the plug on its most powerful AI

The U.S. government on Friday ordered Anthropic to immediately shut off access to two of its most powerful AI models -- Claude Fable 5 and Claude Mythos 5 -- citing national security concerns. Anthropic announced on X that it has complied, but it made clear it thinks the government got this one wrong. The directive, which Anthropic said it received on Friday at 5:21 pm ET, forces the company to disable both models for all users worldwide -- not just the foreign nationals the government's export control order was nominally aimed at. Access to Anthropic's other models isn't affected. Why does any of this matter? Mythos is Anthropic's most capable AI model, one the company previewed in early April and has kept tightly restricted ever since because of what Anthropic described as its exceptional ability to find security vulnerabilities in software. According to Anthropic, Mythos identified flaws in every major operating system and web browser it tested, so rather than release it broadly, the company launched a controlled program called Project Glasswing, sharing it with roughly 50 vetted organizations, including Amazon, Apple, Google, Microsoft, and CrowdStrike, to use for defensive cybersecurity work. Fable 5, released just three days ago, was Anthropic's answer to the obvious commercial pressure: a version of Mythos fitted with guardrails that block responses in high-risk areas like cybersecurity and biology, making it safe enough for general release, the company argued. It was immediately the most capable AI model available to the public, according to benchmark tests from Vals AI, a company that tracks AI tech performance. The government's directive is framed as an export control action, restricting foreign national access to the models. But in a lengthy blog post, Anthropic says its understanding is that the underlying concern is a claimed jailbreak of Fable 5. So far, the company says, the government has provided only verbal evidence of a "potential narrow, non-universal jailbreak" -- one that, as Anthropic describes it, amounts to prompting the model to read a specific codebase and identify software flaws. And by the way, adds the company, it's a "level of capability" that's already widely available in other publicly accessible models, including OpenAI's GPT-5.5. It's also used routinely by cybersecurity professionals for defensive purposes, Anthropic observes. Anthropic's broader argument is that its strongest safeguards operate through independent classifier systems that function separately from the model itself, meaning that even if someone convinces Fable to keep talking past a refusal, the underlying protections against the most dangerous outputs remain in place. The company also notes in it post that a review of recent usage found no evidence of those safeguards being successfully bypassed to produce true, harmful content. Clearly, none of that was enough to stop the government from acting, and Anthropic isn't hiding its frustration. "We disagree that the finding of a narrow potential jailbreak should be cause for recalling a commercial model deployed to hundreds of millions of people," the company wrote. "If this standard was applied across the industry, we believe it would essentially halt all new model deployments for all frontier model providers." Anthropic is widely expected to pursue an IPO this year and has staked much of its public identity on being the safety-conscious alternative to its rivals. The irony isn't lost on observers that the very caution Anthropic displayed in restricting Mythos -- which it promoted as a model so dangerous it couldn't be released publicly -- has now apparently attracted exactly the kind of government scrutiny that could disrupt its business most. OpenAI's Sam Altman must be enjoying this, at least. In April, he told podcaster Ashlee Vance that Anthropic's handling of Mythos amounted to "fear-based marketing." "It is clearly incredible marketing to say, 'We have built a bomb. We were about to drop it on your head. We will sell you a bomb shelter for $100 million,'" Altman said. Altman, whose company is also widely expected to pursue an IPO as soon as possible, didn't predict a government shutdown, but he identified something that has come back to bite Anthropic for now, which is that when you spend months telling the world your AI is uniquely dangerous, the world -- the U.S. government included -- tends to listen.

Anthropic Says It's Taking Claude Fable 5 Offline to Comply With US Government Order

Anthropic says it's disabling two AI models it launched earlier this week, Claude Fable 5 and Mythos 5, to comply with an export control directive it received Friday afternoon from the US government citing national security concerns. The unprecedented incident marks the latest flashpoint between Anthropic and the Trump administration. While the company says the order asked it to suspend access to "any foreign national, whether inside or outside the United States, including foreign national Anthropic employees," it has removed access for all of its customers to ensure compliance. Earlier this year, Trump's Department of Defense labeled Anthropic a "supply chain risk" after the Claude-maker sought to draw red lines over how the US military could use its technology. The designation effectively barred government agencies and contractors from using Anthropic's technology. Anthropic responded by filing lawsuits against the Trump administration. On Tuesday, Anthropic publicly released Claude Fable 5, a version of the company's Mythos AI model with safeguards that prevent it from answering questions about cybersecurity, biology, and chemistry. Prior to the public release, which Anthropic said it had conducted in collaboration with the US government, the Mythos Preview AI model had a limited rollout in April. The goal was to give companies and organizations an opportunity to use its powerful cybersecurity capabilities to improve their defenses, and stem concerns that the technology could be exploited by bad actors to develop powerful hacking tools. In a blog post on Friday, Anthropic says it received a letter from the US government at 5:21pm ET. "The letter did not provide specific details of its national security concern," Anthropic wrote. "Our understanding is that the government believes it has become aware of a method of bypassing, or 'jailbreaking' Fable 5," the company added. "We reviewed a demonstration of this specific technique being used to identify a small number of previously known, minor vulnerabilities. These vulnerabilities all appear relatively simple, and we have found that other publicly-available models are able to discover them as well without requiring a bypass." Got a Tip?Are you a current or former Anthropic employee who wants to talk about what's happening? We'd like to hear from you. Using a nonwork phone or computer, contact the reporter securely on Signal at mzeff.88. In the blog post, the company argued that it has implemented strong safeguards to reduce the likelihood of Claude Fable 5's misuse. Anthropic also claimed that the jailbreak the US government found for Claude Fable 5 was narrow, and would not have made an attacker meaningfully more dangerous than they would have been with another AI model. "To date, the government has only given us verbal evidence of a potential narrow, non-universal jailbreak, which essentially consists of asking the model to read a specific codebase and fix any software flaws," the company said in its blog post. "Our understanding is that one potential jailbreak was shared with the government." Spokespeople for the White House and US Commerce Department did not immediately respond to WIRED's request for comment. Anthropic CEO Dario Amodei said in a policy essay earlier this week that he and the company support a fair, structured, and transparent government process that would block the release of unsafe AI models. In the company's blog post on Friday, Anthropic argued that "this action does not adhere to those principles."

Anthropic Pulls Claude Fable and Mythos AI Models After Feds Claim Jailbreak

Expertise Artificial intelligence, home energy, heating and cooling, home technology. Anthropic pulled access to its newly released AI models Mythos 5 and Fable 5 over the weekend after the US government imposed sweeping restrictions on who could use them. The AI lab said in a statement that the federal government told it Friday afternoon that it had become aware of a way of "jailbreaking" Fable 5, bypassing limits that Anthropic had implemented to reduce the risk the model could be misused. When Anthropic first announced Mythos, it released the software only to a select group of government agencies and technology professionals because of its ability to uncover cybersecurity vulnerabilities. The government imposed what are known as export controls on the products, which Anthropic said means it had to suspend access to the two models by any foreign national, whether inside or outside of the United States. The only way the company could do so is by shutting the models down entirely, Anthropic said. Anthropic said late Friday that it disagreed that the jailbreak it was shown warranted such sweeping action, and that such a standard could "essentially halt" new frontier AI models if applied broadly. "As we have stated publicly, we believe the government should have the ability to block unsafe deployments, as part of a statutory process that is transparent, fair, clear, and grounded in technical facts," Anthropic said. "This action does not adhere to those principles." Representatives for Anthropic and the US Department of Commerce did not immediately respond to requests for comment. Anthropic has reportedly sent staff to Washington to negotiate with the Trump administration to sort out the issue. The Wall Street Journal reported that several hours of calls Saturday included Commerce Secretary Howard Lutnick and National Cyber Director Sean Cairncross along with Anthropic leaders. This isn't the AI developer's first conflict with Washington over AI models. The Defense Department declared Anthropic a "supply chain risk" earlier this year after the company insisted on restrictions against the use of its Claude AI models for fully autonomous weapons or mass domestic surveillance. The Pentagon wanted the ability to use the models for "all lawful purposes."

Who decides when AI is too dangerous?

On today's episode of Decoder, my guest is Hayden Field, senior AI reporter for The Verge. Often when Hayden comes on the show, it's because something has gone wrong in the world of AI. Last weekend, that something was a pretty intense mix of Anthropic, the Trump administration, and Anthropic's new AI model, Fable 5. On Friday, not even a week since Anthropic released Fable to the public, the US government said it was imposing export controls on the new model, as well as the underlying Mythos model that Fable is based on. Those controls restricted foreign nationals, even those working for Anthropic in the United States, from accessing these models. Anthropic then took Fable and Mythos offline for everyone, because the company said it was worried it would not be able to restrict access and reasonably comply with the order otherwise. As you might imagine, this is all a giant mess. Hayden actually just published a fantastic play-by-play on The Verge about how this all went down last Friday and the scramble through the weekend from both sides to figure out what exactly happened and how it might get resolved. So I wanted her to come on and just walk me through the timeline and what it all means. The situation is ongoing. As of Tuesday when we're recording this episode, Fable is still offline -- in fact, if you boot up Claude, it tells you right above the chatbox window that "Fable 5 is currently unavailable." Yet as you'll hear Hayden explain, whether Fable comes back online this week or not, the ripple effects of the government's feud with Anthropic have far-reaching consequences for the tech industry and the US's AI regulatory regime. There's also a big irony here, and you'll hear Hayden and me get into that, too: Anthropic has spent years arguing that AI might soon be powerful enough to be dangerous -- and that the government needed to get serious about regulating AI sooner rather than later. Well... now we're here, and Anthropic doesn't love the way it's playing out. And now, everyone -- but maybe especially the Chinese government -- is watching to see whether the United States' AI regulatory approach takes the shape of a serious safety framework. Or whether it's just another weapon for the White House to use against the companies and people that don't bend the knee to the Trump admin. Like I said, it's a real mess. OK: Verge senior AI reporter Hayden Field on the Claude Fable ban and the new AI regulatory landscape. Here we go. This transcript has been lightly edited for length and clarity. Hayden Field, you're a senior AI reporter here at The Verge. Welcome back to Decoder. Thanks. Great to be here. It's always chaos when you're here, Hayden. That's what I've come to discover. It really is. My rule is still in effect, in that it's always a Friday. Stuff always hits the fan on a Friday and this was no different. Particularly when it comes to the government regulating AI, there's a real "what if it all went crazy on Friday, so we all had to think about it over the weekend." And then I think this has been basically true throughout the Trump administration. The scramble to put it all back together on Monday is immediate. Yep, exactly. It's crazy. And the fact that all these talks were happening all over the weekend and then Monday, still no resolution. I'm surprised that things aren't resolved yet, but yeah, lots of drama. Well, this podcast is going to come out on a Thursday. We're recording on a Tuesday. We'll see what happens over the next two days. But I think the big picture here of how the United States government should regulate AI and what that means for American AI companies on the world stage is open. I don't think it's going to be resolved with whatever happens with Fable. Let's try to understand the whole story and then pull those themes out of it, because they're just going to keep coming back around again and again and again, I think with every new model released now. So, let's start at the start. What is Claude Mythos? What is Fable? How do they relate to each other? Great question. I asked Anthropic the same thing when they came out with this like a week ago. Because AI companies are really bad at naming things and it's always very confusing. But the situation is that Mythos is the underlying model that is powering both Mythos 5 and Fable 5. Fable is a new model. They haven't had Fable 1, 2, 3, or 4. They just immediately started with Fable 5, very confusing. But Fable is the watered down, or safeguarded, version of Mythos 5. When they came out with Mythos Preview in April, Anthropic kept really hyping it up. They were saying this is a potential cyber-weapon. It can never be released to the public, at least until safeguards catch up. We're only going to release it to enterprise and governments, or "cyber defenders" is what they called them -- places that need to patch up all their frameworks before bad actors come in and exploit those vulnerabilities. Now, fast-forward to last week and Anthropic released not only Mythos 5, which is the first official Mythos -- before it was Mythos Preview, so we skipped straight to 5 -- then they also released on the same day Fable 5, which was the very first public version of Mythos-class models. So, same underlying framework, but now tons of safeguards built on top to hopefully stop the public from accessing this super dangerous -- by Anthropic's own admission -- model that could mess up everything. So, that's what they released. And they spent a long time hyping this up and talking about how dangerous it was and then it kind of came back to haunt them later. Fable is a Mythos with security guardrails. When it first came out, there was a wave of controversy, right? AI and security researchers found those guardrails were actually preventing them from doing research and seeing what Fable was capable of, because if they tried to push the boundary, it would downgrade them to Anthropic's previous models. Yeah. It became kind of like a meme. People were making fun of how strict the guardrails were when it came out. So, that's the thing that's confusing me, where there's a controversy because the guardrails are so strict and then something happened that made the government think Fable was so dangerous that it needed to be pulled. What happened? I talked to some independent red teamers about this. Before the Amazon white paper was circulating, before this allegation of a jailbreak was public, I talked to some red teamers about how safe Fable 5 was. They were all pretty impressed with how the guardrails held up. I was expecting it to be a little bit easier to jailbreak. They were saying they had tried a lot of stuff and nothing was working. That's not my experience when I usually talk to independent red teamers. They're like, "Yeah, I got it to break here. I got it to break there." In this case, they were saying it held up pretty well. Now, according to a source familiar with the negotiations who was directly involved that I spoke with yesterday, this is kind of the timeline of last week. In the middle of last week, Anthropic was made aware of some research that Amazon researchers conducted. They had uncovered a potential jailbreak, something they were pretty worried about. They sent it over to Anthropic. Anthropic was kind of going back and forth with the Amazon researchers talking about it. They were discussing whether it was really a jailbreak, or whether it was not that big of a deal essentially. Apparently, reports say that Amazon CEO Andy Jassy then got worried about it and called maybe [Treasury Secretary] Scott Bessent, or some member of the Trump administration, on Friday and talked about his concerns. The Trump administration freaked out. They immediately called Anthropic. They send them a message actually and they say, "Look, you need to shut this down in 90 minutes." 90 minutes? Yep, 90-minute ultimatum. And they said, "We need to figure out how to fix this jailbreak that's come to light." Anthropic got on the phone with the Trump administration within 15 minutes of that first call and they were asking for more details. They're like, "Okay, let's talk about this. Are you talking about the research we're already aware of? Are you talking about something different? What is the jailbreak you're talking about?" They're trying to get more details and talk to the administration. I think the administration was also mad that Dario Amodei, the CEO of Anthropic, didn't call back until an hour and 15 minutes after that first call. Anthropic called back 15 minutes in. It took Dario an hour and 15 minutes to call back. So, there was a lot going on here. But by the time they're talking, the 90-minute deadline passes and the Trump administration says, "Okay, that's it. Yep. We're hitting you with export controls." That is when they told Anthropic that no foreign national can use either of these models and Anthropic has to just make sure that that is the case. No foreign national, whether they work at Anthropic or not. If you're, say, Bank of America and you have access to Mythos 5 and you're using it to plug vulnerabilities in your system and you're not a US citizen, you can't use it. Same with a random, just regular individual using Fable 5. Same with an Anthropic researcher who is not a US citizen. So, Anthropic did the only thing it could really do in the situation and they said, "Okay, we got to sideline both these models. We can't in like one minute figure out a way to make sure no non-US citizens are using the models." So, that's what happened. I mean, it was a crazy couple of hours. Anthropic comes out with a statement later that night saying they're trying to work with the Trump administration on this. They spent all weekend virtually meeting with administration officials and then flying three employees out -- who I detail in the piece we just published -- to D.C. to talk to them in person. Notably, Dario was not there in person. He's virtually joining all these meetings. I can confirm he was not at a wellness retreat, which is what some reports say, but they won't tell me where he is. But he's not at a wellness retreat. That's all I know. Absolutely not at a wellness retreat is Anthropic's position. Okay. [Laughs] So, yeah, there's a lot going on, very dramatic. All the sources I spoke with over the past couple of days were saying the whole tech industry and the cybersecurity community in particular are just talking. One source told me, "Beijing is laughing at us right now." It was interesting because some cybersecurity leaders in the space came together to write a public letter and they said, "Look, not all of us believe in regulation, but if you are going to regulate it, this is not the way," which I think is just indicative of this whole situation. Maybe you believe in aggressive government regulation, maybe you don't. There's a way to operate the regulatory efforts that is rigorous and thoughtful and careful and predictable. Nothing about this seems like it was rigorous or thoughtful or predictable. I actually want to start at the beginning. Andy Jassy and Scott Bessent might be smart guys. They obviously have very senior positions in our society. They're not experts in AI. How did we get to Andy Jassy, the CEO of Amazon, relaying concerns about AI security research to Scott Bessent, who does not really oversee AI policy in any meaningful way, and that resulted in a 90-minute ultimatum delivered to a company whose CEO is absent? I don't think your politics should matter. Either this is the worst security breach in history that the CEO of Amazon is going to call the treasury secretary, or this is just a bunch of guys who want to feel important. And I can't quite figure out which one it is. It's a combination of a couple things. One, it's the fact that, like you mentioned, a lot of these guys are not experts in AI. Any jailbreak of this is obviously serious and worth considering, but I think that it's the type of thing that when you don't have direct knowledge of how these systems work, you're going to take a really black-and-white approach. These guys weren't experts. Andy Jassy hears about this from his own researchers. His own researchers, by the way, who had been going back and forth with Anthropic for a couple days on this. They weren't calling the president about it. They're like, "Okay, let's discuss this and figure out what we can do here." The researchers allegedly didn't seem to think it was the end of the world. They're just like, "Oh, we uncovered this, let's discuss." Anthropic was working on it and then Andy Jassy hears about it. He makes this call, everything hits the fan, which makes sense because if you are not an AI expert -- and even if you are an AI expert, maybe you're not an expert on cybersecurity and how exactly these systems work, of course -- so you're going to freak out. It's understandable, but what is really strange is what happened after that. Make this phone call and the administration becomes aware, okay. But what I think is really weird is the 90-minute ultimatum and not giving Anthropic additional details when they said, "Hey, let's talk about it. Are you talking about the vulnerability we're already aware of or are you talking about something else? And can you just tell us about it so then we can talk about it?" They said, "No, 90 minutes, do it or die." And then making it so that no foreign national can access the system ever. That is what's really strange. Just the way all these cards fell is what everyone in the tech industry thinks is just really strange and it's indicative of, as you mentioned, the way that the Trump administration has been operating with tech. A lot of times it seems to be President Trump has one conversation and he's swayed from that most recent conversation. So he's been very anti-regulation, very pro-American AI and wanting to advance and export American AI, which is something that the commerce department is requesting. The Trump admin has a [request for information] out right now. They're all about exporting American AI and advancing and beating China, but then they do this thing that kneecaps everyone with no notice and could very well impact not just Anthropic, but also OpenAI, Google, every competitor in the space that also has what they call a semi-equivalent Mythos product. It seems really disorganized and it's very interesting that this often ends up in a scramble with all these CEOs trying to race to kowtow to the Trump administration without really being grounded a lot of times in science or evidence that really matches what's going on. The level of freak out is not usually matching the evidence there. Amazon is an investor in Anthropic. They're a major partner. Why did Andy Jassy kneecap his own investment in this way? Well, Amazon typically does a lot of red teaming just normally. It's just part of their process. So it makes sense that this would have been tested and then unearthed. But what's interesting is why Andy Jassy called the Trump administration about it. I'm not sure why. Maybe it was a typical call that was already happening and he just happened to mention something offhand. Maybe he was calling them and saying, "Yeah, good thing you did the executive order because X, Y, Z." We don't know. But what is interesting -- and I'm glad you asked this -- is that a lot of the sources I spoke with said that the stuff in the alleged jailbreak and in the Amazon white paper that's been circulating, all of it could be achieved by OpenAI's GPT-5.5 and some other models. That's the allegation, that these skills are not unique to Fable 5, the skills in question as part of the jailbreak. Mythos is obviously more powerful than Fable because it doesn't have safeguards, but when it comes to Fable and the problems that were brought to light by Amazon, it seems like those problems are not unique to Fable and that GPT-5.5 can also achieve those. A lot of people in the industry are asking, "Why Anthropic?" A real element of this is that it's always Anthropic. Anthropic has the most contentious relationship with the Trump administration. They were designated a supply chain risk by the Department of Defense, but the posture here is different from the usual Anthropic drama. Historically, Anthropic has said, "We don't want you to use our tools for X, Y, and Z." And the Trump administration has said, "No, we get to do whatever we want with the AI tools." That's why Anthropic was designated supply chain risk by the Department of Defense. They said, "We have some red lines about how we want to be used in the military context." Pete Hegseth said, "Nope, I'm going to do whatever I want. You have to allow all lawful purposes" -- that's the line -- "and because you won't, you're a supply chain risk." There's now a lawsuit that's being argued about. Here it's the opposite. The Trump administration is worried that the model is too powerful and can be used for too many things, and Anthropic seems to think that that belief is overblown, that the administration is overplaying a thing that they can fix. They just need the time to do it and some rigor and conversation about how effective their fix is. How do you put that together? On the one hand, the Trump administration is anti-AI regulatory efforts. On the one hand, they say, "We get to do whatever we want with the models, make them as powerful as you can." And then here the model is too powerful and has to be pulled before any foreign national can even look at it. It really just speaks to the fact that the administration is trusting these companies to tell it what it should be doing. We've seen that a bunch with any regulation that has happened or has become voluntary. Lots of drafts have been adjusted after the Trump administration talks to AI companies. They already don't like Anthropic. There's already a big trust breakdown between the two from the supply chain risk situation, from the lawsuit, from drawing red lines. I can't imagine that this exact situation would have played out if this had been a competitor. OpenAI has a Mythos semi-equivalent model of its own. Again, no company is really saying theirs is as good as Mythos, but they say it's pretty good. The freakout would have still happened, but I cannot imagine the rest of it would have played out the way it has. The Trump admin would have given them more time. They would have had more of a conversation and I don't think they would have made this foreign nationals rule that has sent shockwaves through the whole industry and the world. One of the sources I spoke with said that he already knew that backup contracts were being signed with non-US companies right now. Open weight models were being deployed on alternative hardware arrangements because now companies everywhere are saying, "Oh wow, we've got to make political risk part of our business plan now. We don't really know what's going to happen with American AI. Let's make some backup plans here," which is not good for the industry. It's just interesting that the Trump administration is so afraid of China overtaking us, yet they're willing to kneecap one of their champions, as it were, in the industry. But I also think it's interesting that the reasoning from the Trump administration that's been coming out day by day seems to differ. Every day they seem to have a new excuse or part of the conversation that's coming to light. First they say, "Oh, well, what if China had access?" A source I spoke with said that's a rumor from a few weeks ago, from way before Fable 5 or Mythos 5 came out and that it was referencing a Chinese telecom company that the administration had previously approved to have access to Mythos Preview and that when the administration said, "Oh, actually, you know what? We changed our mind. Let's take them off," Anthropic immediately took them off and revoked that access. That's an old rumor. We're also seeing the executive order explanation, where rumors have come to light that Anthropic wasn't complying with the Trump administration's executive order. Well, that executive order hasn't gone into effect yet. He only signed it on June 2nd. It's two weeks ago. It takes 60 days for them to figure out which models are going to be covered under it. There is a voluntary partnership with a 30-day pre-release review. It's voluntary. It's funny because I'm not saying Anthropic's totally in the right here at all. But it's just interesting because in this situation, there's so much misdirection and scrambling and just weirdness that it's rare that you see me feeling bad for an AI company. It really is. I am not like, "Oh yeah, they're doing great." All AI companies have their own issues, but in this one case, I'm saying, "Wow, I really don't know if this would have happened to OpenAI or another competitor, after all the drama Anthropic's been going through with the Trump administration." We know the Trump admin can be petty and I think they would've freaked out for any company that was under this situation, but I can't see it playing out the exact same way with a competitor. Is it just as simple as Dario Amodei doesn't go to the White House? He doesn't give Trump gifts? I mean, Sam Altman does press conferences with Donald Trump. Sundar Pichai attended the inauguration. Zuckerberg was at the UFC fight. [Laughs] Notable AI winner Mark Zuckerberg was at the UFC fight trying to buy his way into success. Is it just as simple as there's a personality clash here? That's the way it's been perceived by some reports. That's definitely part of it. Some reports have said Dario doesn't know how to talk to the administration and that he doesn't kowtow as much. He doesn't know how to play the game. That's part of it. The other part of it is that the Trump administration isn't quite sure how they're going to regulate AI and President Trump is being pulled in different directions. Sometimes he feels like it needs to be more regulated. Sometimes he thinks it needs to be less. If he's talking to an AI CEO, it seems like he's like, "Yeah, you know what? I trust them. We're good. Let them go ahead. We got to beat China." In other cases, he sees that the public is going through a pretty anti-AI era right now. There's a rise in AI populism and he's saying, "Oh, well, actually the everyman doesn't like AI. Maybe I should regulate it." It's something we've seen with the administration a bunch over the past however many years, but now it's being represented in AI regulation. It's just there's not really a rhyme or reason right now. I don't think that Anthropic's communications with the administration help. If you don't take their call the second they call, if you're not flying to the White House to have dinner, it probably hurts your relationship a little bit in a presidency that seems to be pretty relationship-based a lot of the time. I would contrast that with the previous administration, which was almost devoid of personality. The Biden administration, notable for a lack of personality and people, was very technocratic and very rules based in a way that I think people often thought was bloodless. But the Biden administration did put forward an executive order that had things like transparency reviews and research requirements for the frontier models. Trump came into office, and he wiped all that out. He said, "We're not going to regulate these companies." And now it seems like we're just headed back there. If you're going to put out an ultra-powerful model, the administration wants control of it. Certainly they will use the bluntest instrument they have in the form of export controls, but we're going to get back to a more fine-grained, more rigorous approach to how these models are deployed and used and who gets to use them. Was it a mistake to wipe out the Biden framework? Are we headed back to that framework? Where do we think this goes now? I could see us going back there because I interviewed a ton of people in the industry at the time, ranging across a ton of different viewpoints, and most of them believed that the Biden era executive order wasn't quite enough. Some people thought it was a good first step and that's why I could see us going back there because it wasn't seen as being extremely aggressive. If anything, it was seen as being a good first step in the direction, but really just one rung on the ladder. That's why I could see the current administration going back there because it wasn't that far. I will say AI companies will lobby against even that little step. They did. They were pretty mad about it. They did. They did deeply. But from the executive order we just saw, I could see it going back there, but a little more voluntary, which, again, all these things usually get watered down. It's like they come out swinging and then they talk to some AI CEOs and they're like, "Oh, okay, let's roll it back a little bit." So who knows, honestly. But the reason I became an AI reporter is because I felt like there was no real regulation in the space, and this was even before ChatGPT. Journalism is a way to shine a light on what's going on in the industry. Still, six years later, we're not seeing a lot of regulation and any that has happened got rolled back. Maybe we'll go back in that direction. I do see us going somewhere there, but maybe not even as much as that first rung. It almost feels like the industry needs to push for a more considered regulatory approach, right? They need to say, "This is chaos. This can't be how we operate. Every new model can't be forced to contend with panicked phone calls on the weekend." What kind of system would you build if you wanted a more rigorous review process here? Is it that they just submit their models to the government and the government says what's dangerous and not? Is it more industry-based? What are the solutions being proposed? That's a good question. There haven't been a lot of really robust solutions proposed recently because of this current administration. They've had a really light touch approach to regulation so no one's really surfacing those. Now that that might be changing, we'll see more robust frameworks proposed. I just hope that they're not voluntary. AI companies already do a lot of voluntary disclosure. If you're going to regulate, do it. There are a lot of different opinions on how light touch or heavy touch regulation should be, but I do think that voluntary frameworks are pretty watered down. You know what I mean? We already have those. So what's next? It's important to remember how the general public feels because this is who AI is really affecting at the end of the day. There's a rise in anti-AI populism and the fact that people are upset. There are data center protests going on everywhere. People are even, in extreme cases, attacking AI CEO's homes, or at least attempting to. There is a lot of disquiet here and people feel like they're getting replaced. They feel like their jobs are getting replaced. I was listening to a Seattle city council hearing recently and a lot of AI engineers were testifying saying that they got laid off because of AI and they were engineers. Both AI leaders and the government often are pretty distanced from what the general public is feeling about this technology and it's important to note. We'll see how that impacts, if at all, regulation that comes out. I don't have high hopes for a lot of regulation under this administration because they can be really sway-ey. They have dinner one time and then things might shift a little bit. To be fair, many politicians are that way, but we'll see. Maybe after this executive order's 60 days are up, we'll see some changes that surprise me. One of the weirder aspects of this is that Anthropic itself is the company that most pushes for regulation, that most sounds the alarm, is the most doomer-ish of all the companies. Jack Clark, one of the co-founders of Anthropic, recently put out research and his conclusion was that, "We believe it would be good for the world to have the option to slow or temporarily pause frontier AI development to enable societal structures and alignment research to keep up with the advance of the technology." Anthropic is way on the front line of saying, "Oh, we might kill everyone. Boy, we should stop this." But they're also on the front line of model capability. By all accounts, Mythos is the most powerful model that exists today. Is Anthropic just getting itself in trouble? Is this just a marketing strategy for them that's backfiring? It's a combination. The marketing speak is definitely backfiring. If you say something's that dangerous and then a researcher unearths something that maybe wasn't quite meant to be possible, then yeah, people are going to freak out. That's why I can see the first thing happening to anyone. I can't see the rest of it the way that the cards fell. But it's partly there, it came back to haunt them. It was a little bit on them in terms of the hype, but I also think that they are trying to be honest in some ways about the risk that this technology can pose. And it would be really bad if a system that can flag high level vulnerabilities in our global infrastructure that we use, like your bank system or a government website, could just be used by bad actors. If it fell into the wrong hands, yeah, that's pretty bad. It comes down to communication. It's important to talk about the risks and not downplay them, but also if you do that and then you make a mistake or something unexpected happens, you have to deal with the consequences. There's a lot to unpack there, for sure. I also think that a lot of companies do downplay risks of different systems. We've seen a bunch of AI CEOs, including Anthropic's, talk about how jobs aren't all going to be replaced, saying it's going to be okay. I always take all this stuff with a huge grain of salt. At the end of the day, whatever AI company you're talking about, whether it's Anthropic or another one, they all have the incentives to turn a profit and make money at all costs -- especially if they become a public company, which they are about to. People like to paint Anthropic one way and other competitors another, but they're all pursuing the same goal. They just are doing it a little bit differently. Is there a version of this where we just treat AI like military technology from the jump, and the public gets a watered down version of the military technology? Because that seems to be the risk profile that the Trump administration is operating within. Like this is fundamentally offensive technology. We can deploy it against our enemies. We can use it to attack the banks. Everyone is worried about the security implications here. Is there a world in which we just start there and then say, "To deploy to the public, to consumers, you have to cut this way down. We're not going to sell F22s to the public, but we can sell Ford Raptors to the public." There's some version of this, right? Where the technology gets cut way down for the public with the bleeding edge just fundamentally for the military. That's what we've seen a lot for decades and it's starting to happen now with Mythos. Fable is the watered-down version and enterprise and governments have access to the more powerful version. There are a lot of different opinions on that. One of OpenAI's biggest marketing pushes right now is, "Oh, Anthropic wants the most powerful AI to only be in the hands of the elites and we want to bring it to the public." So there are a bunch of different marketing pushes right now. Everyone's just using everything to their advantage to paint their competitors badly, Anthropic included. The race is just heating up so intensely that the AI companies at the forefront are doing anything they can to win at all costs. For some, it's because they think that they are the best and most moral winner, which can always be a big issue as we've seen in every sci-fi or fantasy movie. The for-the-greater-good philosophy is typically used by the villain in the story. There's a big race happening right now and I totally agree with Jack Clark that it might be good to slow things down a bit. If everyone could agree -- which is a huge long shot and especially it'd be very difficult to get other countries to agree as well. But we are definitely building the technology much faster than we can think about the implications of it and we have been for a long time. I think it's the first episode of Decoder where Dario Amodei was directly compared to Thanos, so that's great. [Laughs] To be fair, I was thinking about a villain from Harry Potter when I said that -- Grindelwald. Oh my gosh. All right. Well, Dario, if you want to come on and address these allegations, you're more than welcome. I also have to say, it's not just Dario. All the AI CEOs are saying that, all of them. It's crazy. Where does this end? What are the next steps here? There are meetings happening this week. People are going to be hearing this on Thursday. There's a chance Anthropic is just back on the market. There's a chance it's not. But after this, it seems like everybody wants a process. Where do you think this ends? It's very difficult to predict how this administration is going to act, as we saw with the supply chain risk situation. Even though a lot of government agencies were big fans of Claude, I thought that that was going to be rolled back eventually, but they just signed deals with seven other companies and said, "Okay, we're good. Even if you guys were fans of Claude, whatever, here are seven other options for you." It's anyone's guess right now. I do think that the foreign nationals thing is going to be rolled back. I just can't imagine that after all the hubbub that the administration has created about how American AI has to win, the Hegseth memo from January 9th arguing speed at all costs. I just think this administration is so focused on winning in AI that there's no way they're going to continue with that rule. I could see this becoming a personality game over the next few days. Anthropic compromises in some way, they try to communicate better. And then in a couple of days this gets rolled back. But then I've seen some rumors that the Trump administration is trying to acquire a stake in Anthropic and OpenAI, an ownership stake. So there could be a lot of different "resolutions" to that here. We'll see, but I definitely can't foresee this as it exists right now continuing for more than a couple of days. The industry is just aghast at this situation. Hayden, we're going to obviously have to have you back very soon as this comes to a close. Thank you so much for being on Decoder.

US government warned Anthropic that Chinese group had accessed model, but firm 'refused' to fix Fable 5 jailbreak before US export controls -- Anthropic defended its decision by saying the jailbreak isn't serious

Government concerned that the advanced cyber capabilities of Fable and Mythos would be available 'to people who shouldn't have them.' David Sacks, co-chair of the President's Council of Advisors on Science and Technology and the administration's former AI czar, said the U.S. government warned Anthropic that Claude Fable 5 had been jailbroken and that CEO Dario Amodei refused to fix the flaw or pull the model. In a post on X on Saturday, Sacks laid out the administration's account a day after it ordered both Fable 5 and Mythos 5 disabled worldwide. Sacks said the administration issued the export control "reluctantly" after that refusal, that it wants the restriction lifted once the jailbreak is patched, and that "the ball is in Anthropic's court." Sacks claims that a trusted partner of both Anthropic and the U.S. government, testing Fable, came forward with a jailbreak of the guardrails that separate the consumer model from the unrestricted cyber capabilities of Mythos, the model it's built on. He said the administration asked Amodei to fix the bypass or de-deploy the model, and that Amodei declined. Anthropic instead prioritized keeping its consumer model live over safety, Sacks wrote, calling that inconsistent with the company's positioning as a safety-first lab that had itself lobbied for Mythos to be regulated as a cyberweapon. Sacks also moved to separate the action from Anthropic's earlier clashes with the government, writing that anyone tying the export control to those disputes is wrong, and that the administration values Anthropic's technology and sees the issue as easily resolved. A person close to the White House told the news outlet Semafor that Amazon flagged the jailbreak to the government, and that Amazon CEO Andy Jassy had been in contact with the administration about it. Amazon, which has invested billions in Anthropic and supplies much of its cloud computing, didn't confirm the details, with a spokesperson telling Semafor that governments often seek its counsel on security risks and that it doesn't discuss those conversations. This isn't the first time Mythos access has leaked; it happened back in April when unauthorized third parties reached the restricted model using information from a data breach. Anthropic's public position is that the bypass is narrow and non-universal, that it amounts to asking the model to read a codebase and identify software flaws, and that the same result can be produced on other public models, including OpenAI's GPT-5.5. The company has stated that it disagrees with the notion that a narrow jailbreak should necessitate the recall of a model used by hundreds of millions of people. Sacks rejects this, arguing that a bypass enabling operation of a cyberweapon is difficult to define as anything other than serious. Semafor, citing a person familiar with the matter, says that the White House acted partly over suspicion that a China-linked group had accessed Mythos, raising the prospect of the model being reverse-engineered or distilled. An Anthropic spokesperson told the outlet that the White House "didn't raise Chinese access to Mythos in its conversations around the Fable jailbreak," and that Anthropic blocks access to its products from inside China. Anthropic is separately suing the Pentagon following an impasse over the use of its models in autonomous weapons, and has opposed federal efforts to preempt state AI regulation. Follow Tom's Hardware on Google News, or add us as a preferred source, to get our latest news, analysis, & reviews in your feeds.

Anthropic Blocks Fable 5, Mythos 5 Access Amid Government Order

Anthropic is disabling access to its newly released AI models, Fable 5 and Mythos 5, following a directive from the US government which cited national security concerns. Access to any of Anthropic's other, older models, such as Claude Opus 4.8, will not be impacted. The government order directed Anthropic to block foreign nationals, both inside and outside the US, from using the models. The AI firm said it had no choice but to extend the ban to all users as a result. Anthropic said the government did not specify the exact reason for the order, but that it believes the directive may be related to a potential method for jailbreaking Fable 5. Jailbreaking is a technique used to trick AI models into doing things they are programmed not to do, for example, providing instructions for making explosives. Since it first entered limited deployment in April under the codename Project Glasswing, Claude's Mythos model has attracted significant attention for its ability to find vulnerabilities in software and systems, including products such as Mozilla's Firefox and potentially Apple's macOS. So far, Anthropic has restricted access to Mythos 5 to select US-based partners amid concerns about its offensive cybersecurity capabilities if used by bad actors. Fable 5, which was available to the public, uses the same underlying technology as Mythos, but has safeguards in place for high-risk areas such as cybersecurity and biology. Anthropic said the jailbreak method is "narrow" and "non-universal" and that it has only received verbal evidence of its existence. According to the company's official statement, the technique involves feeding code into the model and asking it to review it for errors. Anthropic agreed that the government should "have the ability to block unsafe deployments" of AI models, but said it "disagree[s] that the finding of a narrow potential jailbreak should be cause for recalling a commercial model deployed to hundreds of millions of people." This is not the first time Anthropic has had serious disagreements with the US government. The company was designated a supply-chain risk in May, which prompted legal action. Meanwhile, President Trump called the company "left-wing nutjobs" in a social media post that same month. Microsoft also blocked its staff from using Claude Fable 5 earlier this week, citing concerns about the company's data retention policy.

Why the US government shut down Anthropic's latest Claude AI model

On June 12, artificial intelligence (AI) lab Anthropic suspended access to its latest Claude models, Fable 5 and Mythos 5, which had been released three days earlier. The move came in response to an "export control directive" from the US government prohibiting use of the models by anyone who is not a US national. Mythos is Anthropic's most powerful, or "frontier", model. When first announcing the model in April, the company said it was too good at hacking to release immediately. Instead, Mythos was made available to a handful of organisations (mostly US tech corporations) to use to patch weaknesses in essential digital systems. Fable is the same basic model, but with added safeguards meant to stop it being used for cybersecurity purposes. This is what was released to the public last week - and almost immediately shut down. Anthropic and the Trump administration at loggerheads Since early 2025, Anthropic and the Trump administration have been in escalating conflict. The administration has accused Anthropic of making "woke AI" and called chief executive Dario Amodei an "ideological lunatic". Early disagreements concerned AI regulation and semiconductor export policy. The dispute sharpened when Anthropic declined to let the Pentagon use its models for domestic surveillance and fully autonomous weapons systems. The Department of Defense responded by threatening to designate Anthropic a "supply chain risk", a classification that would have required military contractors to sever ties. Jailbreaks The US government has not yet publicly stated the reason for last week's directive, but Anthropic it says it believes the government became aware of a jailbreak: a method for circumventing the safeguards in Fable that prevent using its most powerful features for nefarious purposes. These safeguards classify user requests as safe or unsafe before passing them to the AI model. When triggered, the safeguards redirect the request to a less powerful model. The government's concern, according to Anthropic, was that the safeguards could be bypassed to extract information useful for cyberattacks. Guardrails for large language models aren't bulletproof. They mostly depend on the model's own capacity to interpret the user's intentions in making a request. Beyond the inherent difficulty of this task, a large online community (which my colleagues and I call the Undersphere) is working hard to circumvent AI guardrails. Anthropic acknowledges that "perfect jailbreak resistance is not achievable for any current model provider". Anthropic says the research behind the government directive appears to have been produced by engineers at Amazon, which is both a rival to Anthropic and a significant investor. But this was not the only relevant jailbreak. Within 48 hours of Fable's release, a researcher using the pseudonym "Pliny the Liberator" published what they identified as Fable 5's full system prompt to X and GitHub repository. The system prompt is a hidden set of instructions that helps determine an AI model's behaviour. It's unclear exactly how knowledge of Fable's system prompt could be used in practice, but it has drawn attention in the Undersphere. A surprise - and an ongoing mystery The deepest problem of making large language models such as Fable secure is that we don't fully know how they work. According to Oxford University economist and machine learning expert Maximilian Kasy, they work much better than they "should". Large language models have billions of internal parameters and are trained on unimaginably vast piles of data using machine learning methods. According to Kasy, we would expect such systems to be "overfitted": good at reproducing patterns in their training data, but bad at generalising to new situations. However, modern systems such as Claude and ChatGPT do seem to be able to generalise. Kasy likens modern AI development to alchemy: successful through trial and error, not yet grounded in systematic theory. As a result, the behaviour of AI models is partly opaque even to their builders. Hard to regulate The opacity of the technology is one key reason it's so hard to regulate. Governments lack independent access to the data, infrastructure and expertise they would need to evaluate proprietary frontier models. The US administration's recent executive order on AI security, published two weeks ago, reflects this realisation. As the administration has realised the power of frontier AI models, it has moved from an initial hands-off posture to asking developers to share their models for review before release. That demand is an implicit admission that the administration does not trust the companies to evaluate, fully and comprehensively, what their own models can do and how they might be misused. The public sees even less, and the consequence is measurable: a survey taken across 25 countries last year found people are, on balance, more than twice as concerned about AI as they are excited about it. The future of AI safety AI is a hugely hyped technology. But there is no doubt it is also extremely powerful and unpredictable. Understandably, this combination is very dangerous. We cannot rely on regulations, as technology will develop more quickly than they can adapt. Nor can we rely on guardrails, as they will be bypassed. We need a governance framework built for that eventuality: one that can predict and address the consequences of failure. Such a framework must be global, participatory, and founded on reciprocal trust. These are things the current US administration has shown little capacity to generate.

G7 leaders discuss 'trusted partners' access to cutting-edge US AI models, sources say

EVIAN-LES-BAINS, France, June 16 (Reuters) - G7 leaders discussed a plan to grant select "trusted partners" access to advanced AI models from U.S. firms such as Anthropic, three diplomatic sources said on Tuesday, potentially opening a path around restrictions on non-American use. Anthropic on Friday disabled access for all users to Fable 5 and Mythos 5, its most advanced AI models. The company made that move after U.S. President Donald Trumpordered Anthropic to block foreign nationals from accessing its most advanced AI models, citing national security concerns. One of the diplomatic sources said a number of country representatives attending the annual summit of leaders ⁠of the Group of Seven wealthy nations discussed the idea of widening access to advanced AI models with U.S. representatives. This was mainly with U.S. Commerce Secretary Howard Lutnick, on the sidelines of the opening G7 summit dinner on Monday in the French lakeside resort of Evian-les-Bains. The "trusted partners" could be countries or companies, said a second source, who declined to be named because the talks were ongoing. An agreement providing broader access to advanced models would allow G7 countries to use the models to develop stronger cybersecurity defences against rivals such as China. A Trump White House official said in a statement that the president's team has "an open line of ⁠communication with our allies, and we remain committed to addressing national security concerns with Anthropic's model." AI executives from Anthropic, OpenAI and Google (GOOGL.O), opens new tab, which are all developing highly advanced models, are expected to attend a working lunch on Wednesday to speak about technology issues, including regulation, AI infrastructure and networks, Reuters previously reported. Anthropic's spokesperson did not immediately respond to ⁠requests for comment. Cybersecurity experts believe Anthropic's Mythos, a model designed to find flaws in computer code, may turbocharge attacks on banks' technology systems. The European Union is seeking access to Mythos to study the model's implications. Prior to Trump's order, ⁠Anthropic had given access to Mythos to select organizations in "more than 15 countries" so they could use the product to scan their computer systems for vulnerabilities, according to a company statement. The organizations included entities ⁠in the healthcare, communications, power and water sectors, according to the statement. The news of the "trusted partners" scheme was first reported by the Financial Times. Reporting by Michel Rose, Andreas Rinke and Julia Payne in Evian-les-Bains, France; Writing by Gabriel Stargardter and Courtney Rozen; Editing by Matthew Lewis and Cynthia Osterman Our Standards: The Thomson Reuters Trust Principles., opens new tab * Suggested Topics: * Artificial Intelligence * Public Policy Michel Rose Thomson Reuters Michel reports on French politics and diplomacy, having covered President Emmanuel Macron at the Elysee since 2017 and the rise of the far right under Marine Le Pen. He also writes about power dynamics in the EU. He previously covered macro-economics and energy. Worked at the Milan, Italy bureau of Reuters during the euro zone debt crisis and at the London headquarters. Michel is a graduate of the London School of Economics and the Sorbonne, and is interested in mental health and social diversity issues.

US Gov asks Anthropic to ban 'foreign national' access to Fable, Mythos

Anthropic has suspended access to its two most capable AI models, Fable 5 and Mythos 5, for all users worldwide after the US government issued an export control directive ordering the company to block access by any foreign national. US Gov: bar 'foreign national' access to Fable 5 The directive, which Anthropic says it received at 5:21pm ET on June 12, cites "national security" authorities and bars access to both models by foreign nationals inside or outside the United States, including Anthropic's own foreign-national employees. The order's net effect, the company says, is that it must disable both models for all customers to comply. All other Anthropic models, including Claude Opus 4.8, are unaffected. The timing is awkward. Anthropic began rolling out Fable 5 on June 9, free to all Pro, Max, and Enterprise customers through June 22. The model handed to millions for free three days ago is now offline for everyone. Fable 5 is the safeguarded sibling of Mythos 5. Both share the same underlying model, but Fable adds the safeguards. Fable blocks or diverts sensitive cybersecurity, biology, and chemistry queries, while the unrestricted Mythos 5 goes only to vetted government cyberdefenders and life sciences partners. In a developer notice, Anthropic said new sessions would fall back to a user's default model or Opus 4.8, existing Fable 5 sessions would end with an error, and Platform requests to Fable 5 would also fail. It told integrators to migrate to other models. Anthropic's read is that the order stems from a reported way to jailbreak Fable 5. It says it reviewed a demo and found only minor, already-known bugs, the kind other publicly-available models are able to discover without any bypass. "To date, the government has only given us verbal evidence of a potential narrow, non-universal jailbreak, which essentially consists of asking the model to read a specific codebase and fix any software flaws," states Anthropic. "Our understanding is that one potential jailbreak was shared with the government." "We are complying with the government's legal directive and are removing access to Fable 5 and Mythos 5 for all users. However, we disagree that the finding of a narrow potential jailbreak should be cause for recalling a commercial model deployed to hundreds of millions of people." "If this standard was applied across the industry, we believe it would essentially halt all new model deployments for all frontier model providers." The company says the capability is widely available elsewhere, pointing to OpenAI's GPT-5.5, and is used by defenders every day. Anthropic maintains the order is a misunderstanding and says it is working to restore access, while promising more details within 24 hours.

Amazon security research reportedly led to the White House's Anthropic Fable ban

According to the Wall Street Journal, the export control directive that led to Anthropic cutting off access to Fable 5 and Mythos 5 was triggered in part by cybersecurity research from Amazon and conversations between CEO Andy Jassy and the White House. According to the report, the paper from Amazon claims that, through a series of prompts, it was able to get Fable 5 to serve up information that could be used in cyberattacks. Amazon has yet to respond to a request for comment. Shortly after Jassy shared the company's findings with the government, it made the call to block its use by foreign nationals. Complicating this issue is that many of Anthropic's researchers are foreign-born, meaning they were barred from accessing their own product. In a statement, Anthropic disputed the government's characterization of the issue as a "jailbreak." It argued that many of the same vulnerabilities could be discovered using other publicly available models, including GPT 5.5. Some security researchers appear to back the company's interpretation. Katie Moussouris, the founder and CEO of LutaSecurity posted on BlueSky that "I've seen the paper. It's not a jailbreak." Former Commerce Department official Kate Koren speculated to the WSJ that the White House's dislike of Anthropic may have influenced the decision. Anthropic and the Trump administration have been at odds for some time over the company's refusal to allow its AI to be used for mass surveillance of Americans or to power lethal autonomous weapons. In February, Trump instructed federal agencies to stop using Anthropic's AI. And just hours later, Secretary of Defense Pete Hegseth designated the company a supply chain risk. The government and the company seemed to have made amends, and the two had worked together to expand access to Mythos. However, now the two seem destined to clash again.

U.S. gov't orders Anthropic to disable its newest AI models worldwide due to security threats -- ban on Claude Fable 5 and Mythos 5 bars access by any foreign national, even its own employees

An alleged jailbreak triggered a national security directive that forced the model to be pulled. Anthropic disabled its two most capable AI models, Claude Fable 5 and Claude Mythos 5, for every customer worldwide on Friday, after the U.S. government issued an export control directive barring access by any foreign national, according to a statement the company published that evening. Rather conveniently, the order landed at 5:21 pm ET, three days after the models launched, and because it covers foreign nationals both inside and outside the United States, including Anthropic's own employees, the company said selective compliance was impossible and pulled both models globally. The Trump administration's directive specifically targets Mythos-class models, which include Fable 5. Anthropic had released the pair on Tuesday, putting the latter into general availability while keeping the unrestricted Mythos 5 limited to partners in its Project GLassing security program. Both descend from the same Mythos Preview model that Anthropic first announced in April. Commerce Secretary Howard Lutnick sent the directive to Anthropic CEO Dario Amodei, requiring a license for the export, re-export, or domestic transfer of the two models, and extending the restrictions to any foreign person on U.S. soil. With no reliable way to screen foreign nationals out of its user base in real time, Anthropic switched the models off for everyone rather than attempt a partial block. Anthropic said the letter gave no specifics, and that the government has so far supplied only verbal evidence pointing to a narrow, non-universal jailbreak. The technique consists of asking the model to read a codebase and identify software flaws, a task the company said other public models perform without any bypass. It named OpenAI's GPT-5.5 as one model offering comparable capability. "We believe this is a misunderstanding," the company wrote, adding that it's complying with the order while working to restore access. Anthropic also argued that recalling a model deployed to hundreds of millions of users over a single narrow vulnerability, if applied as an industry standard, would halt frontier model launches across the sector. According to Axios, an "administration official" told the publication that the Commerce Department acted after another company claimed it was able to jailbreak Mythos, "alarming the administration about possible national security risks." Mythos is understood to currently be in use by the NSA for offensive cyber operations. The severity of Mythos-class capabilities has been contested since the spring. Independent researchers found that cheaper open-source models could replicate much of Mythos's vulnerability-finding capabilities, and a closer look at Anthropic's headline figures revealed far fewer serious exploits than the marketing implied. Anthropic's relationship with the federal government was already strained before Friday, as the Department of Defense had previously labeled the company a supply-chain risk, and Anthropic has sued the administration over the designation in an ongoing litigation. Meanwhile, the market is already drifting toward open-weight alternatives, most of them Chinese. A March report from the U.S.-China Economic and Security Review Commission found that 80% of U.S. start-ups were using Chinese open-source models, and Chinese labs' share of global model downloads on Hugging Face climbed from roughly 1.2% at the end of 2024 to about 30% a year later. Open-weight families from Alibaba's Qwen, Moonshot's Kimi, Zhipu's GLM, and DeepSeek now hold four of the top five spots on open-weight leaderboards, trailing the best U.S. proprietary models by a margin that has narrowed faster than most forecasts expected: none of them carries a restriction on who can download or fine-tune the weights. Follow Tom's Hardware on Google News, or add us as a preferred source, to get our latest news, analysis, & reviews in your feeds.

Anthropic to meet with Trump administration over Mythos dispute

Anthropic disables new Mythos-class models days after release in response to government directive Senior Anthropic staffers are meeting with Trump administration officials in Washington, D.C., on Monday to try to resolve the artificial intelligence company's latest high-profile dispute with the U.S. government, according to a source close to the company. Anthropic received an export control directive on Friday that cited "national security authorities" and ordered the company to suspend access to its latest AI models, Fable 5 and Mythos 5, "by any foreign national, whether inside or outside the United States," according to a statement. The AI startup disabled access to the models for all of its customers in order to ensure compliance with the directive. The unexpected action marks the latest wrinkle in Anthropic's relationship with the government, which has been strained after a clash with the Department of Defense escalated earlier this year. The DOD labeled Anthropic a supply chain risk in March, which banned defense contractors from using the company's technology because it purportedly threatens U.S. national security. Defense Secretary Pete Hegseth addressed the government's latest directive in a post on X on Saturday, writing that "every passing day" proves why blacklisting Anthropic was "the right move." Anthropic sued the Trump administration in an effort to reverse the supply chain risk designation, and that litigation is ongoing.

Anthropic blocks all customers' access to Fable 5 and Mythos 5 - Engadget

It's to ensure compliance with a government directive citing national security concerns. Anthroic has disabled all of its customers' access to Fable 5 and Mythos 5 in order to ensure compliance with an order it received from the government on Friday, June 12. All its other models and its Claude chatbot are not affected. The company said in its announcement that the US government wanted it to suspend all foreign nationals' access to its newly launched AI models, whether they're inside or outside the US and even if they're Anthropic employees, citing national security concerns. While the US government didn't specify those concerns, Anthropic believes that it's because the government heard about a method of jailbreaking Fable 5. The company has just launched the Fable AI model, which was designed to bring many of Mythos' capabilities to the public, on June 9. If you'll recall, Mythos is its state-of-the-art cybersecurity model that's only available to its Project Glasswing partners. Fable's capabilities "exceed" any previous model Anthropic has launched. It beat Pokémon FireRed during the company's tests, for instance, while Claude failed to beat Pokémon Red, the original game it was based on. Anthropic listed the measures it took to ensure that Fable was secure in its post. It said it instituted strong safeguards to "reduce the likelihood that Fable is misused for tasks related to cybersecurity" and added that its "safeguards are so strong that many users have complained that they are overly broad." The company also explained that any provider cannot possibly ensure perfect resistance to jailbreak attempts, and every model is vulnerable to jailbreaks made especially for it. "We aimed to make jailbreaks either narrow (in the case of non-universal jailbreaks) or very expensive to produce (in the case of universal jailbreaks), and to combine this with thorough monitoring to quickly detect and shut down any successful attacks," it said about its defense strategy. The government apparently gave the company verbal evidence for one potential narrow, non-universal jailbreak that an unnamed entity shared with officials. Anthropic promised to share more details over the next 24 hours, but it clarified that it disagrees that a potential jailbreak should be cause for recalling a commercial model. "As we have stated publicly, we believe the government should have the ability to block unsafe deployments, as part of a statutory process that is transparent, fair, clear, and grounded in technical facts," Anthropic, which has been vocal about its warnings about the need for AI safety, wrote. "This action does not adhere to those principles."

Anthropic to meet with White House officials over AI tool suspension

Bosses at the artificial intelligence (AI) firm Anthropic are set to meet with senior White House officials amid fresh national security concerns over the company's latest release. The meeting is set to take place on Monday in Washington DC between executives at Anthropic and the US Department of Commerce, a government department led by Secretary Howard Lutnick, according to two people familiar with the matter. It comes after Anthropic blocked all public access to the recent release of its latest AI tool on Friday, which it has previously said is "too powerful". The firm made the decision after the US government prohibited Anthropic from allowing any foreign national access to the technology. The AI tool at issue is named Fable 5 or Mythos 5. Fable 5 is a version of the tool with extra safeguards made available to the public, while Mythos 5 has different controls and is only available to a select group of organisations. Both represent a new version of Claude Mythos, a new Anthropic AI model that caused a stir when the company in April initially gave preview and testing access to a relatively small number of organizations, including departments within the US government. Anthropic chief executive Dario Amodei is expected at the meeting with secretary Lutnick, a source said. A White House spokesperson declined to comment. Representatives of Commerce and Anthropic did not respond to a request for comment. By making a version of Mythos available to the public, Anthropic said last week that doing so "comes with risks". "Fable's capabilities exceed those of any model we've ever made generally available", it added. Within days of the release, the US government said it had "become aware" of a potential "jailbreak," or an opening for someone to make an AI tool do something that it was not intended or designed to do. Anthropic said on Friday that it had only received "verbal evidence" of the purported jailbreak vulnerability. The split between Anthropic and the government is the latest this year, which has seen Anthropic sue the US Department of Defense over a fight about how its models can be used. However, tensions appeared to be on the wane a few weeks ago as the company met senior White House officials in what was described as a "productive" meeting. The Monday meeting with Department for Commerce is expected to include more documentation of the alleged issue, according to one of the people familiar. It is unclear, however, if after the meeting Anthropic will be able to make Fable 5 and Mythos 5 accessible again.

Trump Administration Reignites Its Feud With Anthropic Over Latest A.I. Models

The feud between Anthropic and the Trump administration escalated again on Saturday after the government took the unusual step a day earlier of demanding that the artificial intelligence company cut off foreign access to its latest models, as top officials suggested the dispute was unlikely to resolve quickly. Late on Friday, Anthropic disclosed in a blog post that it had disabled access for all customers to its most advanced A.I. systems, known as Fable 5 and Mythos 5, after receiving a directive from the administration to suspend access to any foreign national. The move shocked former U.S. officials and cybersecurity experts, many of whom questioned the validity of the action and noted that it diverged from the hands-off approach to policing the booming A.I. industry that President Trump had endorsed earlier this month. Anthropic said the directive did not explain the national security concerns that prompted it. But the company added that the government had said that it became aware of a method to "jailbreak," or bypass, security restrictions on Fable 5 intended to limit a customer's ability to abuse the product for hacking or other potential harms. Anthropic countered that the fears about the jailbreak method were overblown. The announcement inflamed tensions between the Trump administration and Anthropic, which earlier this month confidentially filed for an initial public offering following a funding round valuing it at nearly $1 trillion. The two sides have sparred for months over how Anthropic's A.I. systems could be used in military and intelligence settings, culminating with the Pentagon labeling the company a supply-chain risk. On Saturday, top Trump administration officials and allies of Mr. Trump's stepped up their attacks on Anthropic. "Three months ago, @DeptofWar kicked @AnthropicAI out of our building -- forever," Defense Secretary Pete Hegseth said on X. "Every passing day proves why that was the right move." David Sacks, a venture capitalist who until recently worked in the administration as an A.I. czar, accused Anthropic in a lengthy social media post on Saturday of being reckless with the release of its latest model, dubbed Fable 5. Mr. Sacks, who said he had spoken to many people inside and outside the administration about the directive, said the administration had asked Dario Amodei, Anthropic's chief executive, to fix the jailbreak issue after a "highly credible trusted partner" of Anthropic's and the government's came forward with research. "Dario refused," Mr. Sacks said. A person briefed on Mr. Amodei's conversations disputed the account, saying that Anthropic was happy to discuss the concerns. Mr. Sacks did not name the "trusted partner" he referred to in his post. But multiple technology firms, including Amazon, spoke with the White House about the security issues, according to people familiar with the matter. Several of the people said a message from Amazon's chief executive, Andy Jassy, detailing security issues was the most influential, raising concerns about the capabilities of the new Anthropic model. But several of these officials said a separate document from Amazon explaining the security concerns with Anthropic's model was misleading. The concerning capabilities that the document highlighted with Anthropic's model are also present in OpenAI's top model, 5.5. An Amazon spokesman declined to comment on the exact nature of its discussions with the White House. "It's not uncommon for governments to seek our counsel on potential security risks," the spokesman said. "When they occur, we don't share the details of these discussions." Administration officials called Anthropic officials at 1:15 p.m. Friday and gave them 90 minutes to pull their most advanced models down, citing an undefined national security concern, according to people briefed on the discussions. Anthropic officials asked for more information and worked to learn what the precise concern was, since the Commerce Department's review and testing of Fable did not reveal significant concerns. Then, at 5:21 p.m., Anthropic was notified that the Trump administration was imposing export controls that effectively forced the company to pull down its model, which consumers had just begun to have access to. Discussions about resolving the dispute are continuing. Howard Lutnick, the commerce secretary, spoke to Anthropic officials on Friday, and was set to have another session with senior company officials on Saturday evening, according to people briefed on the discussions and plans. Some experts said the Trump administration was either misunderstanding or deliberately misconstruing what transpired. Katie Moussouris, the chief executive of Luta Security, said on social media that she had seen the research paper that prompted the administration's actions. "It's not a jailbreak," she said, but rather a defensive maneuver designed to limit the misuse of a model. "If national defense is the goal, this is an own goal." Earlier this month, the Trump administration issued an executive order that asked technology companies to voluntarily let the government review their new models before releasing them to the public. But it did not give the government an official role in approving their release. The order had been delayed after a fierce debate in which A.I. companies -- and some sympathetic government officials -- had pushed back against the government interfering in model deployment. The dispute came on the heels of Anthropic's very limited release of Mythos, an A.I. model so good at discovering and weaponizing new cybersecurity vulnerabilities that it sparked widespread concerns about its potential to wreak havoc. The new restrictions on Anthropic have raised questions about the implications for other A.I. models that may have similar capabilities. Former officials and technology experts said the administration did not appear to have thought through the longer-term ramifications of such a move. But the measure could be limited to Anthropic, which received a fusillade of attacks from the administration in recent months. In February, amid the Pentagon's clash with Anthropic, Mr. Trump called the company a "radical left, woke company" and "Leftwing nut jobs" working to dictate how the government wins and fights wars. "I am directing EVERY Federal Agency in the United States Government to IMMEDIATELY CEASE all use of Anthropic's technology," the president wrote on social media. "We don't need it, we don't want it, and will not do business with them again!" Some administration officials have in recent weeks been looking for an off-ramp in the dispute, U.S. officials say. White House and intelligence officials have pushed forward a classified contract between Anthropic and the National Security Agency, which would allow the spy agency to use the company's technology for a variety of purposes, including intelligence analysis and detecting new computer vulnerabilities. Officials at the N.S.A., which is responsible for digital eavesdropping and government cybersecurity, were not involved in Friday's decision, according to people familiar with the matter. Many U.S. officials said the new technology Anthropic had developed was too important to national security to allow the dispute with the Pentagon to block cooperation. But inside the top ranks of the Pentagon, officials remain upset with the company, insisting that the firm sign on to the same contractual provisions that other A.I. companies have embraced. In a June 12 letter viewed by The New York Times, Mr. Lutnick told Mr. Amodei that a special license would be required for the company to distribute its Mythos and Fable 5 models "to all destinations worldwide," as well as to share them with non-U.S. citizens. "Failure to comply will result in prompt criminal and civil penalties, as provided for by law," Mr. Lutnick wrote. The government has controlled A.I. models before, but past restrictions were more targeted. A measure introduced early last year by the Biden administration placed restrictions on companies' ability to export the so-called weights for specific A.I. models -- the proprietary numerical values that tell the model how much importance to place on different pieces of data. The measure taken by the Trump administration goes far beyond that, by barring Anthropic from sharing the model with any foreign country or any foreign national without first obtaining a license. The restrictions on the involvement of foreign nationals, even those with green cards, could be particularly chilling in an industry that relies on foreign talent. In the past, government officials have often reserved that type of restriction for the most sensitive technologies, like weapons systems. Chris McGuire, a former Biden administration technology official who is now a senior fellow at the Council on Foreign Relations, said that the administration's current policy is that A.I. chips can be exported to China, but Canadian green card holders in the United States cannot access leading U.S. models.

Early users of Anthropic's Mythos still have access after US order

A US shutdown of Anthropic's most powerful model was not total. Who kept access traces the export rule, not merit. When Washington ordered Anthropic to shut off access to its most advanced AI model, it did not shut off everyone. Some of the organisations Anthropic had chosen early to test Mythos have kept their access to a preview of the system, even as other versions went dark under a US export directive, according to a Bloomberg report. The mechanics matter here. Anthropic had limited what it calls Mythos Preview to roughly 200 organisations, including the US government, under its Glasswing programme, after the model demonstrated an unusual capability: it identified thousands of software vulnerabilities. A less powerful version of Mythos was released more widely, then disabled when Commerce ordered Anthropic to suspend access for all foreign nationals on national-security grounds. The preview, for some, survived the cut. Two named firms confirmed they retained access. Dragos, the industrial-cybersecurity company that Accenture is now acquiring a majority stake in, and Cisco Systems both told Bloomberg they still had Mythos Preview. That detail is telling: the organisations keeping access are precisely the ones using the model for defensive security work, the use case Anthropic has put at the centre of its argument for why frontier AI in vetted hands is a public good rather than a hazard. The contrast with Europe is sharp. The European cybersecurity agency ENISA, which had been invited to join Glasswing before Washington's block, was told on Friday it would no longer be given access. That reverses the arrangement that was the subject of a meeting only days earlier, and it answers, bluntly, the question that meeting had left open: whether a European agency could be squared with a US export order. The answer, for now, is no. What the episode exposes is how much discretion sits with Anthropic in the gap the directive leaves. The export order targets foreign nationals, but it does not appear to dictate, organisation by organisation, who among existing Glasswing members keeps a preview running. It was not immediately clear how Anthropic was determining individual access, which means a company is, in effect, making case-by-case calls about who can use one of the most capable security models in existence. That is an uncomfortable position for any vendor, and a revealing one. Anthropic has spent the year warning about the risks of advanced AI while arguing that models like Mythos should be available to defenders. The directive forced it to choose, and the pattern of who kept access, US security firms in, a European agency out, traces the contour of the export rule rather than any judgement about merit. The model at the heart of it is genuinely powerful, which is the whole problem. A system that can find thousands of vulnerabilities is as useful to an attacker as to a defender, and that dual-use quality is exactly why Commerce moved to restrict it. The same capability that makes Dragos and Cisco want continued access is the one that makes Washington nervous about who else might get it. The split-screen outcome also sharpens a question about who governs access to dual-use AI. When a single company decides, in the space left by a government order, which security teams keep a frontier model and which lose it, the practical power over a national-security-relevant tool sits with a private vendor rather than a public authority. That is a lot of discretion to vest in one firm, and the inconsistency between the US firms retained and the European agency dropped makes the discretion visible in a way that invites scrutiny. Anthropic has not detailed its access criteria, and the situation remains in flux as the company navigates between its government's directive and its own customers. What is clear is that the shutdown was not total, that the survivors are concentrated among US defensive-security users, and that the first European to be let in has now been the first to be shown out.

Anthropic's most advanced AI models could be restored shortly following dramatic emergency freeze

Reports allege the freeze stemmed from US concerns over South Korean partner SK Telecom being included in "Project Glasswing" due to alleged ties to China. Last week, the US government sent a shockwave through the AI industry sector by issuing an emergency national security directive that effectively forced Anthropic to pull its most advanced Fable 5 and Mythos 5 AI models globally. Now, the company is striking an optimistic tone, reassuring users that access to its flagship Fable 5 and Mythos 5 models could be restored shortly. Speaking at a press conference in Seoul, Mr. Chris Ciauri, Anthropic's Managing Director of International, stated that the company is "very confident that in the coming days, the models will become available again," as reported by Korea JoongAng Daily. The event in South Korea was initially intended to mark Anthropic's expansion in the region. However, questions about recent export controls and Anthropic's Project Glasswing program (which grants access to Mythos 5 only to select entities) dominated the discussion. The executive declined to comment on most questions related to the shutdown. The US government's security anxiety centers heavily on Mythos 5. Unlike more consumer-focused models, Mythos 5 has been heavily guarded due to its advanced capability to scan, identify, and fix severe vulnerabilities in highly complex, supposedly unbreakable codebases. In the hands of a bad actor, that same capability can be reversed to discover and exploit critical infrastructure flaws. To manage this risk, Anthropic restricted Mythos access to an exclusive corporate cybersecurity initiative called "Project Glasswing." Project Glasswing has approximately 150 partners, according to Anthropic, including Google, NVIDIA, Microsoft, and Apple. However, Wired reports that the US's export controls materialized following a spat with Anthropic over including South Korean telecom giant SK Telecom in Project Glasswing and granting it access to Claude Mythos 5. Citing "people familiar with the matter," the report notes that US officials were concerned about what they alleged were SK Telecom's ties to China. Anthopic declined to comment on Wired's report, while the White House and SK Telecom did not respond. An SK Telecom representative told South Korea's The Chosun Daily, "The anonymous insider's remarks in foreign media lack verified facts, and our company has no ties to China." For paying customers currently locked out of Anthropic's most capable systems, the focus now shifts to whether the company's incoming regional guardrails and security patches will be enough to appease Washington and get the models back online soon enough.

Anthropic Employees and Industry Experts Allege Trump's Model Meddling is Targeted and Unreasonable

The battle between the Trump administration and Anthropic AI has been raging for months, and its mercurial nature has made it one of the harder spats of this term to keep track of, even for those of us in the tech news world. But after last Friday's surprise de facto ban of the company's latest model releases, Fable 5 and Mythos 5, many are wondering if the government's meddling goes beyond the realm of hypervigilance into more malevolent territory. In fact, according to recent reporting in The New York Times, some Anthropic employees are outright convinced that this administration is actively trying to harm their business. The Dario Amodei-led company was initially brought into Trump's orbit late last year with a DoD contract offer as part of the Department of War's rollout of an "AI-first" military strategy. The apparent beginning of things going off the rails came when Anthropic asked for a modicum of regulation and some guarantees about how its tech would be implemented -- mass domestic surveillance and autonomous killings were off the table. The negotiating hurdles presented by these crumbs of ethics were clearly a break from the more laissez-faire deals made by their industry peers and rankled Pete Hegseth. Soon after, in an unprecedented and seemingly punitive measure, the Pentagon deemed Anthropic a "supply-chain risk." This designation barred any "contractor, supplier, or partner that does business with the United States military" from doing business with Anthropic, according to Hegseth's subsequent post on X. Despite this alleged security risk, Hegseth also made clear that Anthropic would still need to spend the remaining months of its contract providing service to the Pentagon until a "better and more patriotic" replacement was found. While both teams' lawyers duked out the legality of the blacklisting in court, Anthropic continued developing its models and, gradually, its relationship with the administration became less adversarial. Or so it seemed until the June release of Fable 5, Anthropic's publicly available version of the Mythos 5 model that powers Palantir's Pentagon efforts, albeit with cybersecurity guardrails. But mere days after that launch, the company was forced to disable both its Fable 5 and Mythos 5 models after yet another edict from the White House. Citing vague concerns about jailbroken versions of the models getting into the hands of "foreign nationals" resulting in a threat to national security, the federal government issued what is called an "export control directive" to Anthropic at 5:21 PM (ET) on Friday, demanding they prevent any foreign access to the models in question. But with paying customers around the globe and so many Anthropic employees themselves foreign nationals, the company's only way to ensure full compliance was to acquiesce to the demands and remove everyone's access, rather than sift through its users one by one. Anthropic says that they were just as shocked by their users about the government directive and maintain that the ban was prompted by overstatements of narrow, known vulnerabilities -- none that come close to warranting such an intervention. Before any re-releases occur, the administration wants to see model versions where guardrails are patently unable to be circumvented. According to reporting in WIRED, a number of experts doubt this hypothetical impregnable model is even possible and believe all cybersecurity guardrails are mere stopgaps waiting to be improved upon once their inherent weaknesses are inevitably exposed. But while the opposing forces continue to meet for negotiations that the President says are "going fine," an ugly precedent is being set. Administrations have always picked horses, but the bullying of Anthropic because they dared to ask for a fraction of the regulation the AI industry desperately needs is another thing entirely. If defending against foreign threats is the real goal here, why have the administration's recent measures against Anthropic compelled employees from their rival, Open AI, to come to do their defense? More importantly, why do these decisions supposedly made for the benefit of the nation always seem to leave America in a more isolated, precarious state than before?

US orders Anthropic to halt foreign access to its most advanced AI models

June 12 (Reuters) - Anthropic said on Friday it has been ordered by the U.S. government to suspend access for all foreign nationals to its Fable 5 and Mythos 5 artificial intelligence models, citing national security concerns. The company said it received the ⁠export control directive on Friday from the government, which gave no specific details of its national security concern. It is Anthropic's understanding, however, that the government believes it has become aware of a method of bypassing, or "jailbreaking," Fable 5, according to the ⁠company's statement. "The net effect of this order is that we must abruptly disable Fable 5 and Mythos 5 for all our ⁠customers to ensure compliance. Access to all other Anthropic models will not be affected," Anthropic ⁠said. Anthropic added that it believed there was a "misunderstanding" and that it is ⁠working to restore access to the models as soon as possible. Reporting by Ananya Palyekar in Bengaluru; Editing by Edmund Klamann Our Standards: The Thomson Reuters Trust Principles., opens new tab

Lawmakers demand answers on the administration's Anthropic restrictions

A bipartisan group of House lawmakers demanded answers from the Trump administration about why it imposed sweeping restrictions on Anthropic's latest artificial intelligence models -- and whether rival tech companies should expect similar treatment. "While this action concerned a single AI model, it appears to represent a significant new application of export control authorities to advanced AI systems and therefore raises important questions for the broader U.S. AI ecosystem, American competitiveness, and the future development and deployment of frontier AI technologies," they wrote in the letter sent Thursday. The Commerce Department placed export controls on Anthropic's newest AI models, named Mythos and Fable, last week, essentially forcing them offline. But the government has provided no official explanation for a move that has shaken trust around the world that topflight American AI systems can be depended upon. The Commerce Department did not immediately respond to a request for comment. It told Anthropic it was imposing the restrictions under powers allowing it to control the export of technology that officials believe could fall into the hands of a foreign military or spying agency, according to a copy of the letter published by Bloomberg. Anthropic's AI models targeted by the administration are closely related. Mythos has been shown to be adept at finding security holes in computer code and Anthropic announced in April that it would provide access to only a small group of partners to avoid harmful use of its technology. Fable is a version of Mythos with additional safeguards designed to tame its hacking powers that was made publicly available last week. Amazon alerted the government that its researchers found there might be a way to bypass those measures, which Anthropic said appeared to have triggered the government ban. The company has disputed the seriousness of Amazon's findings and its technical experts met with government officials this week. The letter from House lawmakers was signed by California Reps. Sam Liccardo (D), Ted Lieu (D) and Jay Obernolte (R), as well as Rep. Scott Franklin (R-Florida). They asked Commerce Secretary Howard Lutnick for information on what risk his department saw in allowing Anthropic's models to remain online and raised concerns about the implications of the sweeping restrictions for the development of AI. Liccardo represents a district that includes much of Silicon Valley. Obernolte, a video game developer, has become a leading Republican voice on AI issues, releasing a proposed package of regulations for the technology this month. Lieu leads a Democratic commission on AI; Franklin was appointed to a bipartisan House task force on AI in 2024. The lawmakers want answers by the end of next week about what analysis officials conducted before imposing the export controls and whether the capabilities of Anthropic's system are different from other AI tools. OpenAI, the maker of the GPT chatbots, has also developed an AI model with powerful cybersecurity abilities. (The Washington Post has a content partnership with OpenAI.) "It is important for Congress, industry participants, and the public to understand what principled distinctions, if any, the Department is drawing among advanced AI models, how those distinctions are evaluated, and what guidance developers can rely upon to assess whether their own systems may become subject to similar restrictions in the future," the lawmakers wrote. Anthropic had clashed with the Trump administration before last week. In the winter, it was involved in a stand off with the Pentagon about rules for the use of its systems by the military. Dario Amodei, Anthropic's chief executive, sought guardrails for mass domestic surveillance and fully autonomous weapons, prompting officials to brand it a national security risk. In recent weeks, the Trump administration also grew concerned about a South Korean telecommunications company that Anthropic granted access to Mythos. Officials considered imposing export controls at that time but Anthropic revoked the company's access and the administration held off on the restrictions, The Washington Post previously reported.

Anthropic disables access to Fable 5 and Mythos 5 to comply with government directive

Anthropic on Friday announced it's disabled access to its Fable 5 and Mythos 5 artificial intelligence models to comply with an export control directive from the U.S. government that cited "national security authorities." The company said it received an order at 5:21 p.m. ET, instructing it to suspend all access to the models "by any foreign national, whether inside or outside the United States, including foreign national Anthropic employees." Anthropic abruptly disabled the models for all of its customers in order to ensure compliance, but said all of its other models will not be affected.

Anthropic's Claude Fable 5 and Mythos 5 AI suspended over security fears

Anthropic has suspended its powerful new AI model after US authorities raised security concerns just days following its public release. In a statement published on its website, Anthropic said it was ordered to suspend foreign nationals from using Claude Fable 5, a program that the company self-described as "too powerful". "The net effect of this order is that we must abruptly disable Fable 5 and Mythos 5 for all our customers to ensure compliance," the company wrote. Anthropic and the Trump administration are involved in a separate ongoing lawsuit over an order to stop government agencies using the company's AI tools. The BBC has approached the US National Security Agency for comment. Claude Fable 5 is a version of Anthropic's Claude Mythos, an AI program rivalling competitors OpenAI's ChatGPT and Google's Gemini. Anthropic said US national security authorities had not identified specific concerns. "Our understanding is that the government believes it has become aware of a method of bypassing, or 'jailbreaking' Fable 5," the company said. Jailbreaking is a process of getting past software restrictions designed to protect a cyber network, allowing hackers to access sensitive information or unblock features. "We reviewed a demonstration of this specific technique being used to identify a small number of previously known, minor vulnerabilities," Anthropic said. "These vulnerabilities all appear relatively simple, and we have found that other publicly-available models are able to discover them as well without requiring a bypass." Ahead of the release of Claude Fable 5, the company touted various "safeguards" it had implemented to prevent cyber hacking. Finance, technology and government leaders had expressed concerns about its public rollout, following a private release in April for previewing and testing vulnerabilities within its own system. Anthropic said it enabled pre-release access for a handful of organisations because the tool was so intelligent that it could be dangerous because of its ability to exploit or hack computer systems. The company self-proclaimed that it was "too powerful to release" before Claude Fable 5 was publicly released, which some critics questioned as inflated hype and marketing spin. "Fable's capabilities exceed those of any model we've ever made generally available," the company said. Anthropic has found itself in the crosshairs of the Trump administration recently. Donald Trump has criticised the company publicly and then US Defence Secretary Pete Hegseth labelled it a "supply chain risk" - the first time a US company has ever publicly received such a designation. The designation means a tool or service is not secure enough for government use, and is historically reserved for companies based in adversarial countries. Anthropic is suing the Pentagon over the designation. A US judge has ruled the Pentagon's directive could not be enforced, meaning government agencies and organisations working with the US military can still use Anthropic while the lawsuit continues.

Anthropic Blocks Foreign Nationals from Using Its Mythos and Fable A.I.

The San Francisco company said on Friday night that the federal government ordered the suspension of the two A.I. systems, citing national security concerns. Anthropic said on Friday night that the U.S. government had ordered the company to suspend access for all foreign nationals to its Fable 5 and Mythos 5 artificial intelligence models, citing national security concerns. The company revealed the order in a social media post, saying that it "must abruptly disable Fable 5 and Mythos 5 for all our customers to ensure compliance." Foreign nationals who work for Anthropic are also affected by the suspension. Anthropic unveiled Claude Mythos in April, but said it could not share the A.I. system with the public because it could become a powerful tool for hackers looking for ways to break into computer networks. The new A.I. system quickly became a security concern in Washington and Silicon Valley because it could represent what experts had worried about for several years:that A.I. represented a new and far more dangerous threat to computer networks. Anthropic shared the Claude A.I. system with about 40 organizations that maintain critical computer infrastructure so they could use the system to patch security vulnerabilities before hackers exploited them. Reactions from researchers who have had access to it have varied. Some agreed that Claude was a new and alarming threat. But others said that it was more evolutionary than revolutionary, and that they could use it to protect their networks just as hackers could use it to break into them.

US order to block foreign access to Anthropic's top models marks a reversal

Anthropic disabled Fable 5 and Mythos 5 worldwide after a directive it calls disproportionate, the first export control aimed at specific AI models. The US government has ordered Anthropic to bar foreign nationals from its two most capable AI models, and rather than try to enforce a nationality rule selectively across a shared cloud service, the company switched them off for everyone. Anthropic disabled Claude Fable 5 and Claude Mythos 5 worldwide late on Friday 12 June, three days after launching Fable 5 as its most powerful public model. It is, by several accounts, the first export-control measure aimed at specific AI models rather than at chips or the hardware that runs them. The directive barred access by foreign nationals both inside and outside the United States, a scope that made selective enforcement on a multi-tenant service impractical and a global shutoff the path of least resistance. The government's concern, as Anthropic understands it, is a method of jailbreaking Fable 5, bypassing the guardrails meant to keep a model from producing dangerous output. The action followed a jailbreak published on X on 10 June by a well-known figure who claimed to have defeated the model's safety controls. Anthropic says it reviewed the report it believes prompted the directive and concluded the capability shown is widely available from other models, naming OpenAI's GPT-5.5 among them. The company called the response disproportionate. Recalling a commercial model deployed to hundreds of millions of users over a narrow potential jailbreak, it argued, could "halt all new model deployments for all frontier model providers" if it became the template. That is the part that travels beyond Anthropic: a precedent in which the government can pull a launched model on national-security grounds, applied across the industry rather than to one firm. The jailbreak at the centre of the dispute was published openly on X on 10 June by a prolific figure in the model-breaking community, who claimed to have got past Fable 5's safety controls. The government's order followed within days, which is part of what Anthropic objects to: a publicly circulated demonstration, rather than a private finding, used as the basis for pulling a model from hundreds of millions of users. For the administration, the move fits a more hands-on posture towards an industry it has otherwise courted, and the willingness to reach into a flagship product days after release is itself the signal. The mechanism also matters. Because the directive targets foreign nationals on a shared cloud service rather than a discrete export, the only practical way to comply was to switch the models off for everyone, which turned a nationality-based restriction into a global outage. For enterprise customers who had built around Fable 5 and Mythos 5, the immediate problem is more concrete: the models they were using are dark, and the timetable for their return is not something Anthropic controls. What happens next runs through Washington.

Anthropic Leaders Are Reportedly in Washington, D.C. Trying to Resolve Their Latest Issues With the Trump Administration

A "source close to" embattled AI company Anthropic says senior company members are working to come to an agreement with the Trump Administration in Washington, D.C. on Sunday after a White House order essentially forced Anthropic's flagship product to be taken offline, according to Axios. On Friday, Anthropic's most prized advanced AI models were disabled by the company after the federal government issued an export control order requiring that Anthropic not allow them to be used by non-U.S. nationals. Claude Fable 5 and Claude Mythos 5 are so-called "Mythos-class" models, meaning they're models built on the same core technology as Claude Mythos Preview, the notorious model Anthropic announced in April. It claimed at the time that Mythos Preview was too powerful to release due to high potential for abuse that may have jeopardized global cybersecurity -- as well as presenting other supposed dangers. Fable 5 and Mythos 5, the models taken down Friday are meant to be tamed versions of Mythos Preview, made safe by extensive and conspicuous guardrails. Those models had a brief life as publicly available products that could be accessed by users with paid Claude accounts, which lasted from Tuesday until Friday night -- supposedly because the federal government was alerted to potential jailbreaks that could have allowed users to bypass the safeguards. What exactly led to the takedowns is slowly being pieced together after the fact. It seems that Amazon was among the companies that drew the White House's attention to supposed potential security holes in Anthropic's Fable 5 and Mythos 5 models. Earlier reporting from Axios claims that on Thursday evening and Friday morning, Amazon and five other companies notified the White House that they had demonstrated the ability to perform the dreaded jailbreaks. According to an additional anonymously-sourced report from Semafor, the White House was also concerned that "a China-linked group" had gained access to a Mythos-class model (Semafor notes that Anthropic doesn't allow access to its AI models from China). On Friday afternoon, Axios claims Anthropic received a call from the White House telling them that keeping Fable 5 and Mythos 5 online constituted an unspecified "national security threat" and that the company had 90 minutes to disable them. It's not clear what happened over the next several hours, but on Friday evening the export control order was apparently delivered via a letter to Anthropic, and several hours after that the models were taken offline. Axios's Sunday story also says the Trump Administration has been unsatisfied with Anthropic's handling of the jailbreak complaint. Anthropic reportedly has not "engaged in a serious manner." Sunday is President Trump's 80th birthday, and the day of a scheduled UFC match to be staged on the White House lawn where a UFC octagon fighting ring has been set up.

"They screwed us": Personality clashes sent Anthropic's models offline

Why it matters: Governing the world's most consequential technology is coming down to speaking President Trump's language. * Anthropic failed to "honor" a recent cyber executive order, administration officials claim, and the company's purported failure to take the matter seriously led to its most powerful products being scrubbed from the internet. * "Everybody said Anthropic was a bad actor. Some of us said it was time to give them a chance. Now those people are questioning that. They screwed us," an administration official said. Catch up quick: On Thursday, Amazon CEO Andy Jassy called Treasury Secretary Scott Bessent expressing concerns that Anthropic's most powerful models, Mythos and Fable, could be jailbroken. * The administration official said Anthropic knew a jailbreak could happen and chose to distribute it anyway: "They came to every fork in the road and took the wrong fork." * Anthropic says it received explicit approval from the government to deploy Fable. * On Friday night the government imposed stringent export controls that ultimately led Anthropic to take the models offline entirely. Behind the scenes: "Anthropic has not done a great job at trying to speak to the administration and appreciate the ideological differences," one source familiar with the administration's thinking said. * "It's like they just speak in different languages," the source said, adding that the company has simply not figured out how to communicate with this administration. The administration first threatened Anthropic with export controls a couple of weeks ago after learning that its cutting-edge Mythos model was made available to an entity in a foreign country with direct ties to the Chinese Communist Party, according to the White House. * A source close to Anthropic said the company has always worked closely with the government on expanding Mythos access -- and in this case, involving a global telecom company, Anthropic revoked Mythos access without the threat of export controls. * Amazon's report raised fresh concerns but Anthropic's "position at the outset was no, we're not going to do anything, this is not a real issue," the source familiar with the administration's thinking said. * The source close to Anthropic said the company did not refuse to resolve the issue. Even before this breakdown, a previous fight between Anthropic and the Pentagon also came down in some ways to just not liking the person on the other side of the negotiating table. * A White House official told Axios that the Pentagon fight is completely unrelated -- but Anthropic's inability to communicate effectively showed up in a similar, unhelpful way. * "We never wanted this to happen. Our number one priority is innovation but our hands were tied," the White House official said. * The optics added fuel to the fire. Anthropic came out with a blog post dismissing the Amazon report. Then the company enlisted a cybersecurity expert viewed by the administration as a "radical Democrat," who was then celebrated by Chris Krebs, who Trump just fired. The big picture: Anthropic has been the loudest of the frontier AI labs on safety concerns, calling for strong regulation and spooking the Trump administration and the public with their own model's cyber capabilities. * The White House led in thawing relations with the embattled company following the Pentagon spat. * The technology is moving fast and the government is struggling to catch up, sources said. That -- combined with the personality differences -- led to a blunt instrument being hastily deployed instead of a scalpel. What's next: "The immediate crisis was averted but longterm we have a problem," an administration official said. * The Commerce Department will meet with Anthropic senior tech staffers Logan Graham, Dave Orr and Nicholas Carlini on Monday, officials told Axios. * Meetings are also scheduled with the CIA and White House science advisor Michael Kratsios to work through adhering to that cyber executive order. The bottom line: One option is to make sure Anthropic's models can't be jailbroken -- though perfect jailbreak resistance may be impossible.

Anthropic blocks all public access to Claude Fable 5, Mythos 5 following US government order -- what enterprises should do

The US government last night issued an unprecedented export control directive ordering Anthropic to immediately suspend all access to its top-tier Claude Fable 5 and Claude Mythos 5 models for foreign nationals, citing unspecified national security authorities. In response, Anthropic has blocked all public access to both models, globally -- meaning no users around the world can access them at this time, even paying enterprise customers and Anthropic employees internally. It's a huge blow and reversal following the public release of Fable/Mythos 5 just three days prior. Current Fable 5/Mythos 5 sessions will end in errors and new queries will be automatically routed to older, less capable models like Opus 4.8. Anthropic says in a blog post that "We believe this is a misunderstanding and are working to restore access as soon as possible," and apologizes to its customers. The sudden regulatory intervention serves as a stark warning to the enterprise sector: centralized, cloud-based frontier models exist at the absolute mercy of government oversight and vendor compliance. Did Pliny the Liberator's public jailbreak catalyze the extraordinary USG action against Fable/Mythos 5? The government's sweeping action follows a viral jailbreak of Fable 5 published publicly on X on June 10 by the prolific jailbreaker "Pliny the Liberator," who claimed to have successfully bypassed the model's safety guardrails to extract functional instructions for cyber exploits, explosives, and chemical synthesis pathways, specifically noting the "birch reduction method" for methamphetamine. Pliny outlined a highly sophisticated, multi-agent attack that leveraged a combination of "Unicode, homoglyphs, Cyrillic," long-context reference tracking, and a technique of breaking harmful requests into innocuous, out-of-distribution tokens. The attacker then used a previously jailbroken Opus model to piece the benign chunks back together into actionable, restricted outputs. Anthropic doesn't specify if this is the jailbreak that precipitated the government order, and in fact, notes that the information provided by the U.S. government regarding the specific jailbreak has been poorly documented, writing: "To date, the government has only given us verbal evidence of a potential narrow, non-universal jailbreak, which essentially consists of asking the model to read a specific codebase and fix any software flaws. Our understanding is that one potential jailbreak was shared with the government." The company argues the capabilities uncovered are "widely available" in other public models, explicitly naming rival OpenAI's GPT-5.5. Furthermore, Anthropic warns that pulling a commercial model over a non-universal jailbreak sets a regulatory standard that could "essentially halt all new model deployments for all frontier model providers". The Pentagon precedent and need for enterprise AI redundancy This sudden blackout triggers severe operational paralysis for organizations relying exclusively on the Claude API. As I warned earlier this year when the Pentagon abruptly blacklisted Anthropic, enterprises can no longer afford -- from an operational reliability standpoint -- to run critical workflows on any single AI model or even provider. Putting all your AI "eggs" into one basket, so to speak, creates a single, ultimately brittle failure point from which recovery or mitigation becomes exceedingly difficult. In March 2026, Secretary of Defense Pete Hegseth labeled Anthropic a "supply chain risk" after the company refused to allow the military to use Claude for mass domestic surveillance and lethal autonomous weapons without safety restrictions. The resulting fallout led to a sweeping prohibition on Anthropic's use across defense supply chains, stripping contractors of access overnight. The lesson from the Department of Defense fallout remains critically relevant today. Any organization building agentic workflows or production apps tied solely to a single closed-API provider risks immediate operational failure if that provider faces an injunction, a cyberattack, or an export control directive. As an enterprise technical leader, your top goal if not already achieved should be to urgently diversify your AI supply. Enterprise implications: sovereign setup vs. frontier capabilities The community reaction to the Fable 5 takedown reflects a rapidly shifting enterprise calculus toward hardware sovereignty. AI founder Alex Finn took to X to flag the Anthropic shutdown as a "wakeup call," urging developers to run local models on home GPUs to insulate themselves from regulatory volatility. "No company or government will EVER be able to take away your local models," Finn writes, warning that government overreach will only escalate as models inch closer to artificial general intelligence (AGI), the stated goal of OpenAI and some other AI firms, in which an AI model becomes capable of performing most economically valuable work tasks now done by humans. Competitors are already capitalizing on this sentiment; Chinese open source AI provider MiniMax quickly highlighted the open weights/open source availability of its new, frontier-class M3 model, contrasting its decentralized availability against Claude's centralized vulnerability. In other words: enterprises can download and run M3 on their own hardware now without ever worrying about any government stepping in to prevent access. This dynamic presents a complex trade-off for CIOs and IT leaders: * The Sovereign Advantage: Running local, open-weights models on sovereign hardware provides absolute control, ensures data privacy, and immunizes the enterprise against abrupt government export controls, vendor policy shifts, or API rate limits. * The Frontier Sacrifice: Adopting a purely local strategy means sacrificing the cutting-edge reasoning, agentic capabilities, and massive context windows inherent to the latest closed-API frontier models, which require centralized, multi-billion-dollar compute clusters to operate. The most resilient path forward is an active fallback architecture. Enterprises must design their systems to be model-agnostic. By building intelligent routing layers that can dynamically switch from a frontier model like Fable 5 to an open-weights fallback or a secondary provider's API the moment an outage or regulatory ban hits, businesses ensure their operations survive the volatile intersection of AI scaling and government oversight.

Anthropic to disable its most advanced AI models after US order limiting foreign access

Company said US government believes safeguards can be bypassed and product used to identify software vulnerabilities Anthropic said it will "abruptly disable" its most advanced AI models for all users after the US government ordered it to suspend access to the models for foreign nationals, citing national security concerns. The company received the export control directive to suspend access to Fable 5 and Mythos 5 for all foreign nationals, without being given specific details of the national security concern, Anthropic said in a statement. It is Anthropic's understanding that the government believes there is a method of bypassing, or "jailbreaking", a safeguard that would prevent Fable 5 from being used in identifying software vulnerabilities, the company said. The order comes just as a previous dispute between Trump administration officials and IPO-bound Anthropic showed signs of easing across parts of the US government. Anthropic's relationship with the government ruptured this year after it refused to allow the US military to use its AI models for domestic surveillance and fully autonomous weapons systems. The government responded by putting Anthropic on a supply chain blacklist, set to take effect later in the year. The action also marks a major escalation of US efforts to halt foreign adversaries' AI capabilities. For years, US export controls have focused on the chips and tools that power AI rather than on restricting foreign access to AI itself. Anthropic said the government has given it only "verbal evidence of a potential narrow, non-universal jailbreak". "We disagree that the finding of a narrow potential jailbreak should be cause for recalling a commercial model deployed to hundreds of millions of people," the company said. The government directive and Anthropic's response highlight growing tension between AI developers and regulators over how to assess risks from so-called "jailbreaks", or methods used to bypass model safeguards. As recently as Wednesday, Anthropic had called for greater US oversight of AI, including the ability to block models with unacceptable risks. It said, however, the government action on Friday did not follow principles of fair and fact-based regulation. The Pentagon's chief information officer, Kirsten Davies, said in a post on X that the defense department supported prioritizing national security. "Some things are simply more important than revenue cycles, clickbait and pre-IPO valuation. America First. Always," Davies said. Anthropic confidentially filed for a US IPO last month, edging ahead of rival OpenAI in the race to reach public markets. Earlier this week, Anthropic rolled out an AI model named Claude Fable 5, representing a new tier of capability it calls "Mythos-class". The model is accompanied by guardrails barring its use in risky areas such as cybersecurity, which some users have complained are "overly broad", Anthropic said. Experts have said that Mythos models, in the wrong hands, could dramatically accelerate sophisticated cyber-attacks, particularly in sectors such as banking that rely on complex, interconnected and often decades-old technology systems. Anthropic said it had worked with the US government, among others, on safety ahead of the Fable launch and that models from rival AI providers showed a similar ability to unearth minor bugs in code. "The net effect of this order is that we must abruptly disable Fable 5 and Mythos 5 for all our customers to ensure compliance. Access to all other Anthropic models will not be affected," Anthropic said. Anthropic said that it believed there was a "misunderstanding" and that it is working to restore access to the models as soon as possible. A US official confirmed that the commerce department had issued an export control directive to suspend all access to Fable 5 and Mythos 5 by foreign nationals.

Cybersecurity executives urge the Trump administration to ease restrictions on Anthropic AI models

SAN FRANCISCO -- A group of cybersecurity executives and experts is asking the Trump administration to lift its directive preventing the use of Anthropic's latest artificial intelligence models by foreign nationals, saying the move could help U.S. adversaries more than it hurts them. Anthropic said Friday it has taken its latest artificial intelligence models, known as Fable 5 and Mythos 5, offline to comply with the directive. The AI giant said it did not believe the steps taken by the government were warranted by the concern it flagged about a potential security issue. Anthropic has said it was limiting use of some its latest technology to select customers because of its ability to surpass human cybersecurity experts in finding and exploiting computer vulnerabilities. The San Francisco-based company has had discussions with the White House previously about the latest models' capabilities. In the letter Sunday, more than 100 cybersecurity experts and leaders from companies including Adobe and Nvidia asked the U.S. government to lift the export control directives on the Anthropic models and "commit to an open, scientific and transparent process of handling AI risk assessments in the future." The letter said that while Anthropic's Mythos models are "quite good" at finding flaws in software and weaponizing exploits, they are "not uniquely good at these tasks" and many of the letter's signatories regularly use other foundation and open-source models for security audits and training. The letter said it is dangerous to take away the best cyber defense capabilities "without a good reason" when America's adversaries are rapidly advancing. China's models, the letter said, are "only months behind the best American models," and it is even likely that China's government has access to private capabilities beyond what's been made public. The export controls marked the U.S. government's most significant step yet to restrict access to the most advanced AI models. Anthropic released Fable widely last week. That model is a limited version of the more advanced Mythos, to which the company has tightly limited access due to cybersecurity fears. The Commerce Department did not immediately respond to a request for comment on Monday. Friday's directive came 10 days after President Donald Trump signed an executive order to establish a framework for the federal government to vet the national security risks of the most advanced AI systems for up to a month before their public release. Participation by AI developers would be voluntary, the order said. Tensions have been running high between the Trump administration and the safety-conscious Anthropic, which has sought to put guardrails on the development of AI to minimize any potential risks and maximize its economic and national security benefits for the U.S. After a contract dispute with the Pentagon, Defense Secretary Pete Hegseth sought to declare Anthropic a supply chain risk, an unprecedented move against a U.S. company that Anthropic has challenged in two federal courts. The company said it wanted assurance the Pentagon would not use its technology in fully autonomous weapons and the surveillance of Americans. Hegseth said the company must allow for any uses the Pentagon deemed lawful.

The week that changed AI: Inside Trump's Anthropic crackdown, and how a phone call from Amazon CEO Andy Jassy triggered the chaos | Fortune

The moment that triggered the Trump administration's dramatic crackdown on Anthropic, and may completely reset the ground rules for U.S. regulation of AI, happened almost by accident. And it was sparked by one of Anthropic's top investors: Amazon. Last week, researchers at Amazon were busy stress-testing Anthropic's newly released Fable 5, a "safe" version of Anthropic's Mythos AI model. Anthropic had repeatedly said that its Mythos-class models had superhuman software hacking skills that were too dangerous to be released to the general public. Fable 5, which Anthropic launched on June 9, was equipped with what it said were robust safeguards against cybersecurity risks. The Amazon researchers testing Fable, however, discovered a "jailbreak" that they documented as allowing users to bypass safety rails and access information that could be used in a cyberattack. Amazon promptly notified Anthropic. But the situation suddenly escalated when Amazon Chief Executive Andy Jassy happened to be on a pre-scheduled call with White House officials last Thursday, June 11, regarding an unrelated topic, according to two sources familiar with Amazon. Jassy brought up the vulnerability his team had found, and White House officials encouraged him to flag the issue directly to Treasury Secretary Scott Bessent, the sources said. Jassy spoke to Bessent that same day. He told the treasury secretary about the Fable jailbreak but also stressed that he was concerned about the cyber capabilities of all of the frontier AI models, including those from other labs. Bessent has been leading the administration's response to Anthropic's Mythos model, largely because of the threat Mythos-powered cyberattacks could pose to the global financial system. His phone call with Jassy set in motion a chain of events that quickly became international news. By Friday evening, just four days after Fable's launch, U.S. Secretary of Commerce Howard Lutnick had hit Anthropic with unprecedented export controls, forcing the lab to pull its most powerful models. The move marked the first time the U.S. government stepped in to explicitly limit the release of a frontier AI model, triggering a wave of tech and political alarm around the world. Since then, cybersecurity experts, AI policy thinkers, American officials, and foreign governments have fiercely debated the consequences, questioning whether the Trump Administration's heavy-handed decision sets a dangerous precedent, or if it reflects wise caution. The stage for the watershed decision had been set months earlier, as the Trump administration wrangled with Anthropic over a separate issue involving its contract with the Pentagon, a fight that deepened a growing animosity toward the San Francisco AI company within the circle of tech policy officials in Trump's orbit. It was further primed by the surging unease within national security agencies as Anthropic began expanding access to Mythos to an broader roster of companies. For Anthropic, which recently closed a $65 billion funding round at a valuation of $965 billion and has filed paperwork to list its shares on the public markets, the export controls represent an existential threat at a time when rivals OpenAI, SpaceX, and Google are competing aggressively to dominate the AI business. Many of the details of the weeklong showdown between Anthropic and the Trump administration are still emerging, and efforts to resolve the situation are fluid. Fortune spoke to more than a half dozen people involved on various sides of the situation to piece together how it happened and what it could mean going forward. An ultimatum, and a flurry of frantic calls On Friday, following Jassy's phone call with Bessent, the administration contacted Anthropic and Chief Executive Dario Amodei, demanding that they either fix the jailbreak or remove the models from the market. According to a source familiar with Anthropic, the warning arrived with no prior communication or indication from the government that it considered the export of Fable a national security threat. Despite the lack of warning, the Commerce Department issued a 90-minute deadline, followed by strict export controls banning foreign nationals from using the software, a directive that Anthropic interpreted as an effective ban on the models themselves. That's because the government uses a "deemed export" standard that interprets the distribution of an AI model to any foreign national, including those physically located in the U.S., as the equivalent of exporting the model. That standard would apply even to Anthropic's own non-U.S. employees. What's more, Anthropic had no effective way of verifying the nationality of its users. The ultimatum set off a flurry of frantic calls between Anthropic and the White House. Anthropic maintained that the jailbreak was relatively simple and could easily be achieved using other available models, the source said. But the argument failed to sway administration officials, who maintained that the bypass unlocked dangerous cyber capabilities. An official letter from Lutnick, whose department oversees export controls, arrived just after 5 p.m. Eastern Time on Friday, informing the company that it required government approval before making its Fable 5 and Mythos 5 models available to any user abroad or any foreign national globally. Lutnick warned that non-compliance could result in criminal and civil penalties, according to a copy of the directive obtained by Bloomberg News. By 10 p.m. that evening, Anthropic had taken Fable 5 and Mythos 5 offline. "We disagree that the finding of a narrow potential jailbreak should be cause for recalling a commercial model deployed to hundreds of millions of people," the company wrote in its blog post after taking the models offline. "If this standard was applied across the industry, we believe it would essentially halt all new model deployments for all frontier model providers." An Anthropic spokesperson added that both parties are working quickly to resolve the issue. Representatives for the White House did not respond to multiple requests for comment. The saga continued into the weekend and this week. By Sunday, senior Anthropic technical staff had arrived in D.C. to meet White House officials in person, the source familiar with Anthropic said. "The Admin values Anthropic's technical capabilities and feels that this issue, while serious, should be easily resolved," David Sacks, a venture capitalist who serves in an advisory role in the Trump Administration, wrote on X over the weekend. "The ball is in Anthropic's court." But by Tuesday, Fortune understands, those talks had concluded with the export controls still in place and the next steps unclear. The two sides reached an impasse. Industry blowback While the immediate future of Fable remains in limbo, many cybersecurity experts have come to Anthropic's defense. They warn that government licensing of AI models will make U.S. labs less competitive and allow geopolitical rivals like China to swoop in. "Anything that stops the evolution of technology and development of technology is just going to leave the U.S. behind," Danny Jenkins, chief executive at cybersecurity company ThreatLocker, told Fortune. "Criminals don't care whether they're allowed to export something from the U.S. or not." At a demo this week, Jenkins said he used Anthropic's Claude model, which is not considered as potent as Fable or Mythos, to "completely ransomware a machine," which the model itself eventually identified as textbook ransomware. But, Jenkins said, vulnerabilities are going to be exposed eventually, whether by a human or an AI model. The AI model simply does it faster. He suggested the government's focus should be on hardening security and creating "zero trust" environments, rather than trying to police a model's existence. Legal experts have also questioned the logic of the export controls. Charlie Bullock, a senior research fellow on LawAI's U.S. Law and Policy team, told Fortune that Lutnick's letter leans on authorities that were never designed for publicly accessible AI systems, especially where users only interact with a model via an API (a software interface that lets different programs communicate). That distinction raises questions about whether the government has stretched its statutory powers and whether the move can withstand First Amendment scrutiny, since earlier court decisions have treated published computer code as a form of protected speech. "I think the takeaway is that the export control letter is on very shaky legal ground," Bullock said. Since first flagging the issue to White House officials, Amazon has taken a step back on negotiations between Anthropic and the U.S. government. Amazon is one of Anthropic's largest investors, having invested at least $13 billion in the startup in a series of deals that date back to 2023. One engineer at Amazon said that employees have been making jokes that their company "snitched" on Anthropic, while others lamented the fact that they never had an opportunity to use Fable 5. In a statement, an Amazon spokesperson said that as one of the world's largest cloud providers, "it's not uncommon for governments to seek our counsel on potential security risks. When they occur, we don't share the details of these discussions." While it's unclear whether Jassy expected his comments to White House officials to result in the export controls, it's almost certain that he would have been aware of the tension between Anthropic and the Trump administration. International reverberations Back in October, Sacks, who used to serve as the Trump AI and crypto czar, lashed out at Anthropic for advocating for AI regulation at both the federal and state level, accusing the company of running a "sophisticated regulatory capture strategy based on fear-mongering." He had previously accused the company of being "woke" and staffed with "leftists." This was followed by a highly publicized feud between the lab and the Trump Administration over the Defense Department's use of Anthropic's models. The Pentagon wanted Anthropic to agree to a new contract that would make it clear it could use the company's models for "any lawful purpose," while Anthropic wanted specific exemptions built into the contract that would bar its models from being used to control autonomous weapons or for domestic mass surveillance. When the two couldn't come to terms, the Trump administration ordered all federal agencies to stop using Anthropic's models and labeled Anthropic "a supply chain risk," a designation that had never before been applied to a U.S. business, and which meant Anthropic's models could not be used by any part of the U.S. military or by any defense contractors when working on Pentagon contracts. Anthropic has challenged the designation in court. When Anthropic released Mythos in April, the relationship with the U.S. government appeared to be resetting on better footing. Because Mythos can find previously unknown security vulnerabilities -- including in open source code bases that have been subject to extensive scrutiny by human security researchers for decades -- and can then autonomously string these vulnerabilities together to run entire exploits, Anthropic coordinated with U.S. national security agencies. Anthropic's Glasswing program initially released the model to a carefully curated group of 12 launch partners, all U.S. technology or financial firms whose systems controlled critical U.S. infrastructure. The idea was that these companies would use Mythos to find and patch vulnerabilities in their own software before hackers, who are increasingly using AI to find vulnerabilities and automate attacks, discovered them. Amazon and Microsoft were part of the initial group, which was then expanded to some 40 additional U.S. software companies in consultation with U.S. security agencies. But the model's debut also sparked panic among foreign governments and businesses, who were clamoring for access to Mythos to help them secure their code bases. Several weeks ago, Anthropic gave the U.S. government a list of 111 trusted organizations, many of them located overseas, that would also get access to Mythos. The administration approved this list. However, the lab later informed officials it had expanded the list by about 50 additional organizations, The Washington Post reported. One of those new entries, the newspaper reported, was a South Korean telecommunications firm the administration suspected of having ties to China. A source close to Anthropic said the company did receive a request from the U.S. government several weeks ago to cut off Mythos access to a global telecommunications company. The source said Anthropic did this quickly after receiving the request. Other news accounts suggest the company was SK Telecom, although SK issued a statement saying "there is no basis for considering our company to be linked with China." The revelation of the ballooning Glasswing roster and the inclusion of the Korean telecom firm in that group damaged trust between the U.S. national security establishment and Anthropic, according to the Post. Even before Fable was rolled out last week, the National Security Agency was actively pushing the administration to put emergency export controls in effect, the report said. Can the rift be repaired? Amodei reportedly spent last weekend working the phones, trying in vain to persuade the administration to reverse its decision, while the company prepared to send its delegation of technical experts to Washington. According to press accounts, government officials were insisting Anthropic create guardrails for Fable that can't be bypassed. But many experts in AI security say that all AI models are currently susceptible to some form of jailbreaking and that there is no known technical fix for this vulnerability. The rift between Anthropic and the White House was on display Wednesday on the global stage at the G7 summit in Évian-les-Bains, France, where Amodei and Trump found themselves in the same room for a roundtable discussion of more than two-dozen business and government leaders. Seated directly at the president's side was Sam Altman, Amodei's arch-rival from OpenAI, who has remained publicly mum about the latest feud between the U.S. and Anthropic. Across the room, on the other side of a large oval table, Amodei sat next to French President Emmanuel Macron. Addressing the group, the Anthropic CEO reportedly offered a pointed warning: "resist the temptation to splinter" in how they handle AI regulation. On Thursday, Politico reported that Anthropic and the U.S. government were working on a deal to restore access to Fable and Mythos that would also see them jointly develop a framework for judging when future jailbreaks of a model's guardrails posed a serious danger. Meanwhile, in Seoul, South Korea, Chris Cauri, Anthropic's managing director of international, told reporters at a press conference that "we are very confident that in the coming days, the models will be available again." But, as of Thursday evening Pacific Time, exactly how that would happen remained a mystery.

The US government just hit the brakes on Anthropic's most powerful AI models

Anthropic disables Fable 5 and Mythos 5 after US government directive Anthropic's troubles with the US government do not seem to be easing. The company has now been ordered to suspend access to Fable 5 and Mythos 5 for all foreign nationals, including foreign national Anthropic employees working inside the United States. Anthropic said it received the directive on June 12 and is disabling the two models for all customers to comply. Other Anthropic models are not affected. The government has not publicly explained the full national security concern, but Anthropic says it understands the order is linked to a reported method for bypassing, or jailbreaking, Fable 5's safeguards. A fresh clash after the Pentagon fight This is not Anthropic's first serious standoff with Washington. Earlier this year, the company was caught in a dispute with the Pentagon after it refused to remove restrictions preventing Claude from being used for fully autonomous weapons and mass domestic surveillance. That fight led to claims of blacklisting and legal action, putting Anthropic's safety-first position directly at odds with parts of the US government. The latest directive puts Anthropic back in a familiar position. Officials are worried about access to powerful AI systems, while Anthropic argues that its safeguards are being misunderstood or judged by an unrealistic standard. Why Fable 5 became a concern The concern around Fable 5 is tied to Mythos 5's advanced cybersecurity capabilities. Anthropic has said Mythos-class models can discover and exploit software vulnerabilities, and Mythos 5 was reportedly tested by the NSA and other government-linked evaluators before wider release. While those capabilities can help security teams identify and fix weaknesses, they also create national security concerns if they are used for offensive or malicious purposes. Fable 5 was released only a few days ago as a public version of Mythos 5 with stricter guardrails. Anthropic said it was designed to block or redirect sensitive cybersecurity and biology-related queries to Opus 4.8. Recommended Videos Anthropic says the reported bypass only surfaced minor, already known vulnerabilities and that other public models can do similar things. Still, with a topic as sensitive as cybersecurity, caution is not unreasonable. If Mythos 5 is capable of identifying software vulnerabilities at a high level, then its guardrails cannot be merely good enough. They need to be airtight. Anthropic may argue that the reported jailbreak was narrow, but the government's concern this time is easier to understand. In this case, "better safe than sorry" may be the government's most defensible position.

US order cutting access to Anthropic's AI models sparks criticism

New York (AFP) - The US government's order for Anthropic to withdraw its most powerful artificial intelligence models has sparked a wave of criticism from both advocates and opponents of AI regulation. On Friday evening, the San Francisco-based company announced that the US Department of Commerce had ordered it to suspend Mythos 5 and Fable 5 for "national security" reasons, without providing further details. Unlike Mythos 5, which was unrestricted and available only to a small number of partners, Fable 5 was heavily protected to prevent any major misuse, particularly for cyberattacks or the development of chemical and biological weapons. But Anthropic said an organization -- whose identity it didn't disclose -- reported to the Trump administration that it had found a way to bypass safeguards designed to prevent Fable 5 from being used for a cyberattack. Anthropic described the loophole discovered by this third party -- identified by several media outlets as Amazon -- as "narrow" and said the software vulnerabilities it exposed were "minor." The directive applied only to access by foreign nationals, but Anthropic said it was unable to distinguish among users based on nationality and was therefore forced to take its models offline. A government's outright ban on an advanced AI model developed by a domestic company is unprecedented. China blocks access to the most capable Western AI models and imposes restrictions on major domestic AI companies, but those restrictions are generally built into the models before they are released. An 'impulsive' decision Entrepreneur Martin Varsavsky said the implications of the order are "enormous." Any startup "making frontier models is at the mercy of the government," he commented on X. "Therefore, the order doesn't just punish Anthropic. It changes the rules for the entire industry." Researcher Gary Marcus said he saw the United States and China battling to a "tie" in the AI race -- until Friday's government announcement. "It didn't occur to me the Trump administration could trip the US efforts from behind," Marcus said. "But it just did." Some observers argue that Anthropic bears considerable responsibility for its predicament after it had warned for years about the risks associated with the most advanced AI models. On Wednesday, Anthropic CEO Dario Amodei once again called for policymakers to "activate a slow and rickety policy apparatus to deal with risks and opportunities that are going to compound surprisingly quickly from here." Several of President Trump's supporters who, until only a few weeks ago, strongly opposed AI regulation -- - much like the Trump administration itself -- - have attempted to defend the directive. Among them are influential investor Marc Andreessen and former White House AI adviser David Sacks. Others, however, including former Trump AI adviser Dean Ball, accused them of intellectual dishonesty, noting that they had fiercely criticized regulatory efforts under former president Joe Biden. The pro-regulation group Americans for Responsible Innovation argued that decisions of this magnitude should not be made "impulsively" or be subject to "political favoritism." Anthropic is currently at odds with the Trump administration, which has terminated all of its government contracts with the company. Many observers agree that AI has entered a new era requiring greater government involvement, but they strongly objected to the manner in which the action is carried out. "In a functioning administration, nobody would have ever been blindsided by an action like this," said Ben Murphy of the Institute for Progress, a think tank focused on emerging technologies. "The government simply would have just requested that Anthropic do additional testing or add more safeguards before release," he wrote on X. AI's rapid acceleration and the concentration of influence in the hands of a few companies have caught governments off guard, said Mona Sloane, a professor at the University of Virginia. That means that "it is possible that we will see" government-imposed suspensions of AI models again, she said.

Europe reacts to Anthropic halting access to top AI models

Anthropic said it believed the US government had become aware of a potential means of jailbreaking Fable 5. European politicians have sounded the alarm on sovereign AI after US firm Anthropic announced it was halting access to some of its top AI models following a directive from the Trump administration. The company said late on Friday that it had received a letter from the government ordering it to suspend access to its Fable 5 and Mythos 5 AI models for foreign nationals over national security concerns. "The net effect of this order is that we must abruptly disable Fable 5 and Mythos 5 for all our customers to ensure compliance," Anthropic said. The news has prompted a wave of reaction across Europe, with many politicians saying it should come as a major reality check for governments and calling for increased investment into homegrown tech. Euronews has compiled some of the responses below. Bruno Retailleau, former French interior minister and 2027 presidential candidate "Washington's decision to cut access to Anthropic's most powerful models should serve as a wake-up call. In the race for artificial intelligence, a nation that depends on others for its technology is a nation that can be unplugged overnight," he wrote on X. "France has unique assets in Europe: nuclear electricity that is decarbonized and sovereign, engineers among the best in the world, but also companies like Mistral, OVHcloud, Scaleway, and ChapsVision, capable of competing with American giants. But we must put an end to naivety and decide, at last, to rearm our technological power". Al Carns, British MP and former minister for the armed forces "This week the most advanced AI model on the planet got switched off by a foreign government. British researchers were studying it. British companies were testing it. British hospitals were piloting it. Not any more," said Carns, who quit his role as armed forces minister over a defence spending row earlier this week. "This isn't an AI story. It's the story of every industry we used to lead," he continued. Geert Wilders, leader of the Dutch far-right Party for Freedom "I want my #Anthropic Claude Fable 5 back!," Wilders wrote on X. "AI is more and more national sovereignty," he added, while also calling for the Netherlands to accelerate the development of its own models. Benjamin Haddad, French minister delegate for Europe "Trump administration's decision to bar foreign nationals from accessing Anthropic's latest model marks an accelerator of the geopolitical battle over AI," Haddad wrote. "Europe cannot settle for being an open market dependent on technologies designed, funded, and controlled elsewhere," he went on. "It must invest more, support its innovators, and equip itself with the means to master the technologies that will determine power in the 21st century". Tom Tugendhat, British MP and former security minister "Disabling Fable 5 and other models for foreigners is not a misunderstanding or a mistake, it's the inevitable result of technology shaping warfare so that sovereignty is more about code than cannons," Tugendhat said. He added: "With high energy costs and the emphasis on safety not opportunity Britain's response has been to build the brake cutting ourselves off from the future and tied ourselves to the past. We cannot continue like this and remain sovereign." Édouard Philippe, former French prime minister and mayor of Le Havre "By restricting access to Anthropic's most powerful models for non-Americans, the US government is choosing to subject AI development to its logic of power," Philippe, who served as French PM from 2017 to 2020, said. "AI is now a critical infrastructure, as essential as electricity or the Internet. An infrastructure whose models and computing power we do not control is an infrastructure that others can unplug," he added. Jordan Bardella, MEP and president of France's far-right National Rally party "The Trump administration suspends access to Anthropic's latest AI models for all non-American nationals: this sudden decision comes to remind us that artificial intelligence is already a major issue of national sovereignty," he said in a post on social media. "Nations that do not quickly develop their own model(s) will always depend more and more on the choices of other powers: France must accelerate its support for the gem Mistral AI and the entire AI ecosystem," he added.

Anthropic suspends new AI models after government directive

The Anthropic logo can be seen at an event organized by the AI company. Andrej Sokolow / picture alliance via Getty Images file The U.S. government instructed Anthropic to prevent any foreign national from accessing the AI company's latest and most powerful models Friday evening, according to the company. Anthropic, the company behind the popular Claude chatbot, said this directive forced it to disable access to Fable 5 and Mythos 5 for all customers -- not just foreign nationals -- to comply with the directive. "The US government, citing national security authorities, has issued an export control directive to suspend all access to Fable 5 and Mythos 5 by any foreign national, whether inside or outside the United States, including foreign national Anthropic employees," the company wrote in a statement posted online late Friday evening. The letter issuing the directive was sent from Commerce Secretary Howard Lutnick and was written with the help of officials from the Commerce Department's Bureau of Industry and Security, according to an administration official. Anthropic released Fable 5 and Mythos 5 on Tuesday, trumpeting the new models as the most powerful AI systems it had ever shared. The company said that these AI models are powerful enough to merit strict safety guardrails to prevent malicious or dangerous use. "Releasing a model this capable comes with risks. Without safeguards, Fable 5's capabilities in areas like cybersecurity could be misused to cause serious damage," Anthropic said in its Tuesday announcement. While both models were built on the same technical foundation, only Fable 5 was released to the general public. The Fable model includes much stronger restrictions on the types of questions that users can ask, especially on issues related to cybersecurity and biology. Mythos 5, meanwhile, was released without such safeguards to a select group of trusted partners, including key cybersecurity and infrastructure companies.

Amazon's CEO reportedly triggered the government crackdown that shut down Anthropic's most powerful AI

Amazon's Jassy reportedly told the government that researchers used Fable 5 for cyberattack info. That triggered the order to shut down Fable 5 and Mythos 5. Amazon CEO Andy Jassy was reportedly the source of security concerns that led the US government to force Anthropic to shut down Fable 5 and Mythos 5 for every customer on Friday. The Wall Street Journal reported that Jassy told Treasury Secretary Scott Bessent and other officials that Amazon researchers used Fable 5 to obtain information that could be used in cyberattacks. The government then imposed an export control ban on both models. The revelation adds an uncomfortable dimension to the story. Amazon is one of Anthropic's largest investors, having put in billions and receiving a $100 billion cloud spending commitment in return. The company that bankrolls Anthropic's infrastructure is also the one that told the government its models are dangerous. An Amazon spokesperson said it is "not uncommon for governments to seek our counsel on potential security risks" but declined to share details of the discussions. The spokesperson also pointed to an AWS status update confirming that Amazon's own cloud platform was affected by the model shutdown. David Sacks, Trump's former AI czar who now co-chairs the President's Council of Advisors on Science and Technology, offered a different account. He said "a highly credible trusted partner of both Anthropic and the USG came forward with a jailbreak." Sacks added: "The Admin asked Dario to fix the jailbreak or de-deploy the model. Dario refused." Anthropic's version is different. The company said it reviewed the jailbreak technique and found it surfaced "a small number of previously known, minor vulnerabilities." It called the government's response disproportionate and said the capabilities causing concern are already available in other publicly accessible models. The shutdown affected every customer globally because Anthropic cannot filter foreign nationals from US users in real time. The sequence matters for the AI industry. Amazon invested billions in Anthropic. Anthropic built models on AWS. Amazon's CEO told the government those models pose security risks. The government used export controls to force them offline. AWS was then affected by the shutdown. It is an ouroboros of corporate dependency and regulatory blowback. For Anthropic, the immediate question is how quickly it can restore access. For the broader industry, the precedent is what matters. If a major cloud provider can trigger an export control action against its own portfolio company by raising concerns with the Treasury Secretary, the competitive dynamics of the AI market just gained a new weapon. Mythos is used by banks and government agencies for vulnerability discovery. Every one of those customers lost access because of a dispute between Anthropic's biggest investor and the government its CEO has been lobbying for more regulatory power.

Scoop: Trump admin blocks foreign access to Anthropic's most powerful AI

Why it matters: The move marks an escalation in Washington's effort to treat cutting-edge AI systems as national security assets. * Anthropic now finds itself on a Pentagon blacklist deeming it too dangerous for the government's own use, and in a Commerce Department licensing regime deeming it too dangerous for foreign use. Driving the news: Commerce Secretary Howard Lutnick on Friday sent a letter to Anthropic CEO Dario Amodei saying that the Mythos 5 and Fable 5 models would be subject to export controls to any location outside of the U.S. and to all foreign persons within the country. * An administration official told Axios the Commerce Department decided to take the action after another company claimed it was able to jailbreak Mythos, alarming the administration about possible national security risks. * The administration tried to get Anthropic to pause releasing the latest models but was unsuccessful, the official said, prompting the export control letter. * The model needs to remain locked down until the U.S. government's national security apparatus is hardened, the official said, adding that could happen in the next few weeks. * Anthropic did not immediately respond to a request for comment. Zoom in: Per Commerce's letter, a license will be required for the export, re-export or domestic transfer of Anthropic's models. * Furthermore, Anthropic will have to submit an additional application for individually validated licenses. * Failure to comply would result in financial and civil penalties. Context: The Trump administration earlier this month released an executive order to test the most advanced AI models before they are deployed. * Anthropic has a partnership with the Center for AI Standards and Innovation at Commerce for pre-deployment testing. Yes, but: The executive order is voluntary and explicitly avoids a licensing regime -- something White House chief AI adviser David Sacks was able to secure to avoid what he considers the "regulatory capture" of the biggest labs. * An administration official said that Trump "does not want to hurt the industry and wants innovation to continue." The bottom line: Anthropic's running fight with the government just got more complicated.

Trump keeps kneecapping the U.S.'s most promising AI models

Bullies vs. Brains: Anthropic's new scrape with the Trump administration Anthropic is in another fight with the Trump administration, this time over its new Mythos-class models, released last week. On Friday, administration officials panicked after reports that Amazon researchers had tricked Claude Fable 5 into providing cybersecurity information Anthropic had tried to block. Officials gave the company 90 minutes to voluntarily take Claude Fable 5 and Claude Mythos 5 offline. Anthropic, waiting for evidence that the models had actually been compromised, did not immediately comply. The administration then declared the models a cybersecurity risk and barred foreign nationals from using them. Because Anthropic had no practical way to limit access to U.S. citizens only, it shut the models down for everyone. As a result, nobody can use Claude Fable 5 or Claude Mythos 5, likely the most powerful publicly available AI models in the world. That may keep bad actors from exploiting them, but it also prevents cyberdefense researchers and software companies from using them to stop cyberattacks. Even foreign nationals who work at Anthropic are now barred from using the models.

US bans foreign access to Anthropic's Mythos and Fable

Close aide to Anthropic and US government reportedly raised security concerns around Mythos and Fable. Countries across the world are reeling from the effects of an abrupt suspension of two of Anthropic's most capable AI models following an order from the US government late on Friday (12 June). Now, senior technical staff from the AI giant are reportedly in Washington to meet with White House officials in hopes to resolve their second major dispute in months. In a surprise statement on Friday, Anthropic said that the US government issued an export control directive, ordering the company to suspend access to Claude Fable 5 and Claude Mythos 5 by "any foreign national, whether inside or outside the US, including foreign national Anthropic employees". The order has effectively halted the usage of the company's industry-leading cybersecurity models Mythos and its more restrained version, Fable, released just last week. The letter, sent to Anthropic at 5:21 pm EST (10:21 pm Dublin Time), cited national security concerns but did not provide specific details, the company said. "Our understanding is that the government believes it has become aware of a method of bypassing, or 'jailbreaking' Fable 5. "These vulnerabilities all appear relatively simple, and we have found that other publicly-available models are able to discover them as well without requiring a bypass. "We have reviewed a report that we believe is the basis of the government's directive and validated that the level of capability displayed there is widely available from other models," the statement read. David Sacks, a former AI 'czar' and the co-chair of the President's Council of Advisors on Science and Technology said that a "highly credible trusted partner" of Anthropic and the US government "who was testing Fable" told officials about a potential jailbreak to Fable's guardrails. "In their blog post, Anthropic defended its decision by saying the jailbreak isn't serious. That is not what the trusted partner and the USG believe; nor is that kind of minimising language consistent with Anthropic's brand as the AI safety company," Sacks wrote in a post on X. And Pentagon's chief information officer Kirsten Davies said that the move prioritises national security. "Some things are simply more important than revenue cycles, clickbait and pre-IPO valuation," Davies said. Meanwhile multiple media publications reported on Saturday (13 June) that Amazon CEO Andy Jassy was among the tech leaders who raised concerns with the US government over potential security concerns in Anthropic's models, leading to an export control ban. Amazon is a major investor in Anthropic. A company spokesperson for Amazon told TechCrunch that its "not uncommon for governments to seek our counsel on potential security risks". The European Commission is among the foreign regions facing the ramifications of the sudden suspension of Anthropic's latest models. Several of the bloc's countries, as well as the UK, had been in discussions with the company over Mythos and its potential applications in securing their systems. In a statement on Sunday (14 June), the Commission said that it is analysing the implications of an export control directive. "We are seeing a new generation of highly capable AI models reach the market. These models offer significant benefits, including for cyber-defence, but they ⁠also raise serious cybersecurity concerns that need to be addressed," a European Commission spokesperson said. "We believe that contingency measures ⁠taken in this light should not be discriminatory against partners. "This development is a further ⁠illustration of why Europe needs to strengthen its technological sovereignty...We are looking ⁠closely at the practical consequences of this for European users of these services." In a similar statement, Canadian prime minister Mark Carney said that the incident highlights the risk of overreliance on a small number of powerful AI tools. This is the second major dispute Anthropic has had with the US government, which could have drastic consequences for the AI giant as it gears to go public in an estimated $1trn-plus listing. The company is currently engaged in a legal battle with the Trump administration following a separate incident earlier this year after Anthropic refused to change its AI safeguards for usage by the US government. Anthropic said that its late-February 'supply chain risk' designation by the US government was not legally sound. Its more recent statement last week makes similar connections, claiming that the Trump administration's actions do not adhere to principles of transparency, fairness and facts. The tug of war between the Anthropic and the US government comes as the latter eyes technological domination in AI over its contemporaries. Don't miss out on the knowledge you need to succeed. Sign up for the Daily Brief, Silicon Republic's digest of need-to-know sci-tech news.

Why the moral of Anthropic's fable reads like an unhappy ending for the AI sector far beyond the US

Last week Anthropic unleashed Claude Fable 5 on the world; 48 hours later the US Government cut the world outside America off from the same tech, citing national security concerns - and a wake-up call rang out all around the globe. It's the latest turn in the ever-souring relationship between Trump 2.0 and the US champion that until earlier this year was lauded as the only AI model supplier secure enough for use in the highest levels of the Pentagon, but overnight was deemed a pariah run by "left-wing nutjobs" for refusing to back down on two ethical red lines in its contract Ts & Cs. That led to President Donald Trump decreeing that Anthropic was not to be used in any US public sector work and warning that others who partnered with the firm were also to be excluded. Meanwhile all US Government agencies had six months to clear out any Anthropic dependency they already had. That's a battle that's still making its way though the US legal system, but the conflict took another turn on Friday when the order went out to shut off access to two of its most powerful AI models -- Claude Fable 5 and Claude Mythos 5 -- as well as a ban on non-American nationals from accessing them, This led Anthropic to suspend the model for all users worldwide - and set sovereignty alarm bells off internationally. What happened? According to Anthropic's version of events, the US authorities raised concerns over a jailbreak uncovered by third parties - rumored, but not confirmed to be Amazon, Anthropic's biggest investor! While admitting to "minor vulnerabilities, the firm had pitched Claude 5 as a version of Mythos complete with guardrails in high risk areas and therefore deemed safe for commercial use. It argues that it hasn't been presented with the evidence to justify the US Government's actions: To date, the government has only given us verbal evidence of a potential narrow, non-universal jailbreak, which essentially consists of asking the model to read a specific codebase and fix any software flaws. We validated that the level of capability displayed there is widely available from other models including OpenAI's GPT-5.5, and is used every day by the defenders who keep systems safe. No testers have yet been able to find a universal jailbreak -- a jailbreak method that can very broadly bypass the model's safeguards, unblocking a wide range of cyber capabilities. The firm is complying with Trump 2.0 edict, for now, but making it perfectly clear that it disagrees with the decision: We are complying with the government's legal directive and are removing access to Fable 5 and Mythos 5 for all users. However, we disagree that the finding of a narrow potential jailbreak should be cause for recalling a commercial model deployed to hundreds of millions of people. What comes next? What happens now remains to be seen. Anthropic might try to update the model to prove further that the tech is indeed as safe as the firm claims. Or it might lawyer up again in another action against this fresh US Government ruling. Beyond this clash itself, the knock-on effects of this latest test of strength are going to be felt across the AI sector as a whole and around the world as a fresh indicator of the 'Balkanization' of AI that has been warned about for some time. This ought to serve as a strident warning signal to any country that isn't the US (or China) that dependency on US tech and models is deeply unhealthy. If a US administration can wake up one morning and cut off your access to the tech on which you're building and running mission-critical systems, that's a hugely vulnerable position to be in. It's a sign, say many commentators, that the push for sovereign AI needs to be a priority for all countries. But the push towards that would mean, for example, Europe to ease off on the regulatory stranglehold on AI that it seems hell bent on pursuing and nations like the UK to pull back on their focus on areas such as net zero in order to expand out the infrastructure demands of data centers and lower cost energy that easing off of US tech and onto home grown alternatives would need to meet. Sure enough the usual tut-tutting and 'something must be done' noises are already being made in the usual quarters. In the UK, for example, Kanishka Narayan, Minister for AI and Online Safety, solemnly pronounces: The main lesson: as we debate the future of national security and technological sovereignty, access to AI capabilities is crucial. This is then followed up a shameless, self-congratulatory political tribute to the pittance recently announced by the UK Government to support its stated goal of turning the country into a world-leading AI economy to be reckoned with. File under, 'Something must be done, this is something, I've done something' - and move on. Can anyone recommend a good Chinese model to check out? On the home front... Meanwhile back in the US, how far will this latest Trump 2.0 edict go towards disincentivizing crucial engagement with the government, outside a few of the 'usual suspects' who will forever be 'in the room'. The latest move certainly reeks of an escalation in the politicization of the AI sector, driving from central control. At the most extreme, the government gets to decide who uses AI and which AI, at that. There's a need for regulation, but there are limits beyond which regulation becomes authoritarianism. Of course, maybe there's another reading to all this? Is is possible this is just a revenge strike by Trump 2.0 after being defied by Anthropic earlier in the year? That view doesn't seem that unreasonable when you consider the comment made by Pete Hegseth, Secretary of War, who sneered: Three months ago, [the Department of War] kicked Anthropic out of our building forever. Every passing day proves why that was the right move. It was left to others to offer a more measured political response, including an intervention by David Sacks, co-Chair of the President's Council of Advisors on Science and Technology and the man said to have Trump's ear on all matters to do with AI policy. He says: Keep in mind that Anthropic itself widely promoted the idea that Mythos was a cyber-weapon and needed to be regulated as such. They asked for regulation of Mythos and championed the guardrails on Fable. Of the supposed 'jailbreak' reported by "a highly-credible partner" of both Anthropic and the US Government, he claims: The Admin(istration) asked [Anthropic CEO] Dario [Amodei] to fix the jailbreak or de-deploy the model. Dario refused. And he adds: The Admin's hope now is that Anthropic remediates the safety issue, the export control is lifted, and Fable goes back into general release. The Admin wants all this to happen as soon as possible. It is frankly baffled that Anthropic hasn't wanted to comply with safety requests that it previously said were its highest priority...the ball is in Anthropic's court. One last comment from inside Trump 2.0, this one from Kirsten Davies, CIO at the Department of War, who declared: We fully support [the President] and [the Secretary of War] in prioritizing national security and the security of our warfighters, DIB partners, critical infrastructure, international partners and allies. Some things are simply more important than revenue cycles, clickbait, and pre-IPO valuation. America First. Always. Meanwhile Elon Musk just became a trillionaire on the same day this kicked off. There's a probably a fable we can learn from there as well... My take I'm going to defer to Aaron Levie, CEO of Box, who provided an articulate response to the situation on X: It's an early peek into what AI regulation would end up looking like at scale when enacted at the model layer instead of the specific application of AI. The government would have sole discretion over when a model can be released to the public, based on a bunch of factors that they inherently control. He went on: Thus the whole challenge [as a vendor] will be that you're debating with the government, over months and months, with every model release, what these models are actually capable of and what their risks are. Inherently there's not only a lot of subjectivity in determining those risks, but there's also many other factors that go into the risks being practical in the first place. The net result would be negative for the US AI sector, he concludes: We would end up with a backlog of AI releases, progress in the market inherently would dramatically slow down, and AI would start to look more like any other sclerotic industry. If this paradigm had existed three years ago at the start of the current AI wave, we'd likely currently be stuck on GPT-4 level intelligence at this point. That's not exactly a situation compatible with the Trump 2.0 doctrine of US dominance of the global AI market at any cost. The other conclusion to be drawn here is that Anthropic hasn't read another fable - that of The Boy Who Cried Wolf. As many have pointed out, the firm has demonstrated a propensity to talk about AI models, and its own tech in a way that pushes the limits of transparency and skirts the boundaries of self-flagellation. Just a week or so ago, Amodei published his latest missive, Policy on AI Expontential, in which he warned that current AI regulations aren't enough and government can't keep pace: AI is progressing extremely fast -- much faster than the policy process was built to handle. Amodei's counterpart and one-time colleague Sam Altman over at OpenAI called the Anthropic approach was tantamount to "fear-based marketing", warning: It is clearly incredible marketing to say, 'We have built a bomb. We were about to drop it on your head. We will sell you a bomb shelter for $100 million'. When Sam Altman sounds Iike the one talking sense, it's probably time to re-think your policy...

White House forces Anthropic to disable new frontier models following abrupt export ban

White House forces Anthropic to disable new frontier models following abrupt export ban Anthropic PBC was forced just to abruptly pull the plug on foreign access to two of its most powerful artificial intelligence models, Fable 5 and Mythos 5, just days after it launched them. The sudden decision just before the weekend to do so came shortly after the U.S. Commerce Department issued a sweeping export control order citing urgent security concerns over the models. The Commerce Department order explicitly bans all "foreign nationals," including those of allied nations and Anthropic's own foreign employees from accessing the Fable 5 and Mythos 5 models. Because of the difficulty in restricting usage based on nationality alone, Anthropic responded in the only way it could - by disabling access globally to ensure it could comply with the order. In a statement on X, Anthropic said the order came following a "misunderstanding" by U.S. national security authorities over the risks of the new models. It explained that government officials had provided it with "verbal evidence of a potential narrow, non-universal jalbreak." It's believed that the jailbreak relates to a statement from the Red Team lead at the U.K.'s AI Safety Institute, which said four days ago that its cybersecurity team had made "substantial progress" towards a universal jailbreak of Fable 5. Posting on X on June 9, Xander Davies of the U.K.'s AISI wrote: "Within a few hours of access, we developed a jailbreak that extracted malicious responses to single-turn question-answering." He added: "Over two additional days of dedicated testing, we extended the jailbreak to sometimes allow for multiple steps of malicious agentic tool-calls." It was not immediately clear if the export control was issued as a result of the AISI's jailbreaking efforts. The Wall Street Journal reported that the Commerce Department was, in fact, alarmed by a jailbreak report from Amazon.com Inc., which also had access to the model. According to the Journal, Amazon Chief Executive Andy Jassy personally called Treasury Secretary Scott Bessent and other officials to inform them that his company's researchers used Fable 5 to obtain "information that could be used in cyberattacks." In a post on X, David Sacks, co-chair of the President's Council of Advisors on Science and Technology, explained that the White House had asked Anthropic Chief Executive Dario Amodei (pictured) last week to patch or de-deploy Fable 5, only for the CEO to decline. Anthropic hit back, saying that the flaw discovered by the U.K. team was only a "minor" risk, and also existed in rival models such as OpenAI Group PBC's GPT-5.5. It warned that the U.S. government was applying heavy-handed standards that would "essentially halt all new model deployments" for all frontier AI labs. Anthropic is likely to suffer massive collateral damage as a result of the decision to roll back Fable 5 and Mythos 5. The incident will dent its reputation as a reliable partner in the eyes of its growing enterprise customer list. Moreover, the ban means that many of Anthropic's own employees are no longer able to interact with the technology they helped to create. In Europe, the ban has ignited a firestorm of criticism from America's allies. European Commission spokesperson Thomas Regnier said the EU is currently assessing the practical consequences of the ban. "We are seeing a new generation of highly capable AI models reach the market," he said. "These models offer significant benefits, including for cyber-defence, but they ⁠also raise serious cybersecurity concerns that need to be addressed." "We believe that contingency measures ⁠taken in this light should not be discriminatory against partners," he added. French and U.K. politicians were even more blunt. British MP Tom Tugendhat complained that dozens of British hospitals and companies that had access to the model and were carrying out vital research had their access cut off overnight. He said that the global ban shows how AI is redefining the geopolitical landscape. "Disabling Fable 5 and other models for foreigners is not a misunderstanding or a mistake, it's the inevitable result of technology shaping warfare so that sovereignty is more about code than cannons," he said. Meanwhile, French political figures Jordan Bardella and Bruno Retailleau said the ban is a "wake-up call" for Europeans, and urged them to back homegrown companies such as Paris-based Mistral AI to ensure that their essential digital infrastructure cannot be switched off on a whim by Washington officials. The shutdown exacerbates the already toxic relationship between Anthropic and the Trump administration. Earlier this year, the Pentagon added the company to a supply chain blacklist following its refusal to give the U.S. military unrestricted access to its AI models so it could be used to automate weapons and perform mass surveillance. Anthropic is currently suing the government over the decision to blacklist it. From a financial perspective, the timing of the ban could be disastrous, as it comes just days after Anthropic filed documents with the Securities and Exchange Commission ahead of an initial public offering expected in late summer or fall. The company is believed to be racing to list publicly ahead of OpenAI, and is chasing a $1 trillion valuation. But the ban means it will lose substantial revenue from enterprises. That, combined with the uncertainty of the regulatory environment around AI, is almost certain to dissuade some investors and alter the company's IPO timeline. In the longer term, there is a risk of the U.S. creating a serious rift between itself and its allies. By prohibiting them from accessing its most powerful technology, it increases the likelihood that they'll pivot away from American technology providers and focus on domestic alternatives in order to protect themselves from the kind of disruption this incident has caused.

Anthropic shuts down newest AI model after U.S. bans foreign use

Anthropic CEO Dario Amodei speaks at the AI Impact Summit in New Delhi on Feb. 19. (Bhawika Chhabra/Reuters) The artificial intelligence lab Anthropic said Friday it had shut down access to its newest and most powerful models after the U.S. government banned use of the technology by foreign nationals, citing national security concerns. The San Francisco-based company said it received a letter from the government at 5:21 p.m. informing it of the decision. Anthropic said the letter did not specify officials' concerns, but that it was informed that the government believed there was a method of bypassing safeguards designed to neuter the ability of its Fable model to aid computer hackers. The export restriction prevents any foreign national from using the technology -- whether they are inside the United States or overseas -- including employees of Anthropic. That limitation could upend work inside the company. "We disagree that the finding of a narrow potential jailbreak should be cause for recalling a commercial model deployed to hundreds of millions of people," Anthropic said in a blog post. "If this standard was applied across the industry, we believe it would essentially halt all new model deployments for all frontier model providers." The government action represents a sweeping restriction on AI technology -- a decision all the more remarkable for being issued by the Trump administration, which until recent months had pushed to sweep aside regulations that might slow innovation. It comes after a previous effort by the Pentagon to brand Anthropic a national security risk amid a fight over the use of its technology by the military. Neither the White House and nor the Commerce Department responded to requests for comment on the restrictions. Fable is a version of Anthropic's new Mythos model with additional safeguards built in. The company initially declined to release Mythos publicly, saying it had potentially dangerous abilities to identify security flaws in computer code and help hackers exploit them. The development of the model set off a scramble inside the government to understand the risks, an effort that culminated earlier this month in President Donald Trump signing an executive order aimed at giving officials an advance look at especially powerful new AI tools. Taking part in any reviews would be voluntary, according to the order, which said it did not authorize the "creation of a mandatory governmental licensing, preclearance, or permitting requirement." Anthropic released Fable this week, saying it had taken steps to ensure it was "safe for general use," and some users complained that its guardrails were too restrictive. In its blog post Friday, Anthropic said it had reviewed a report that it believes is the basis of the government's concerns and that the findings did not translate into a hacking risk greater than that posed by some other AI models. The company has called for the government to have the power to block the release of new AI systems if they pose a safety risk. But it said Friday it would require a "process that is transparent, fair, clear, and grounded in technical facts." "This action does not adhere to those principles," the company wrote.

Anthropic disables Fable and Mythos AI models after U.S. government bars it from giving foreigners access | Fortune

Anthropic was forced to disable all access to its newest AI models, Fable 5 and Mythos 5, late on Friday after the U.S. Commerce Department used national security export controls to bar the company from distributing the models to any foreign national. The directive includes not just people located outside the U.S., but also any foreign national in the U.S., including Anthropic's own non-citizen employees. Given the scope of the directive, Anthropic argued it had no choice but to disable the models for all users. It clarified that access to its less powerful Claude models, including its latest Claude Opus 4.8 model, was not affected. "We apologize for this disruption to our customers," Anthropic wrote in a post on X. "We believe this is a misunderstanding and are working to restore access as soon as possible." Anthropic said in a blog post that it received the directive at 5:21 pm Eastern Time. It said the letter it received "did not provide specific details" of the government's national security concern. But the company also said that officials had told Anthropic that the government made the decision after learning of a technique to bypass Fable 5's safeguards. These safeguards were designed to prevent users from accessing the powerful cybersecurity abilities of Mythos, the underlying AI model on which Fable 5 is built. Anthropic said it believed the jailbreak the government was citing was a narrow one that would unlock Mythos's cybersecurity capabilities in only one specific instance and not a universal one that would defeat all of Fable 5's safeguards. It also said it believed the same jailbreak could be used to elicit similar capabilities from other publicly available models, including OpenAI's GPT-5.5, that are not subject to similar national security export controls. "We disagree that the finding of a narrow potential jailbreak should be cause for recalling a commercial model deployed to hundreds of millions of people," Anthropic wrote in its blog post. "If this standard was applied across the industry, we believe it would essentially halt all new model deployments for all frontier model providers. "As we have stated publicly, we believe the government should have the ability to block unsafe deployments, as part of a statutory process that is transparent, fair, clear, and grounded in technical facts. This action does not adhere to those principles," Anthropic said. AI industry insiders and policy experts reacted with disbelief to the unprecedented U.S. directive. Some saw the move as a further attempt by the Trump administration to punish Anthropic. U.S. President Donald Trump in February ordered all federal agencies to stop using Anthropic's models after the company refused to agree to the Pentagon's preferred contract terms for AI vendors, which stipulated that any AI models it purchased could be used "for any lawful purpose." Anthropic had been seeking exemptions from having its models used for autonomous weapons systems or mass domestic surveillance. The Pentagon declared Anthropic a "supply chain risk" in early March, requiring the U.S. military to cease using its models and prohibiting defense contractors from using them for government contracts. Anthropic is challenging that designation in federal court. Several key Trump technology policy advisors, most notably former AI and crypto czar David Sacks and Pentagon undersecretary of defense for research and engineering Emil Michael, have publicly attacked Anthropic and its executives. Sacks has accused Anthropic of being "woke" and "leftist" as well as engaging in "a sophisticated regulatory capture strategy based on fear-mongering." Dean Ball, an AI policy expert who briefly served in the Trump administration but is highly critical of its recent decisions around Anthropic, said on X that "I can't tell if this is lawfare against Anthropic in particular or extreme national-security hawkery. Regardless, it is simply cartoonish." He added: "An administration whose posture is that we *should* export advanced AI chips to China, which also wants to ban... Britain (and every other non-American on Earth)... from using our best models? I have no words." But others said that Anthropic was simply reaping what it had sown. When Anthropic first debuted its Mythos model, the company argued the model was too dangerous to release broadly. When it released its Fable 5 model, which is based on Mythos, the company highlighted the safeguards it had put in place to prevent users from accessing Mythos's full capabilities. "If you describe your product as a munition in every press release, eventually a government takes you at your word," Peter Girnus, a cybersecurity researcher who describes himself in his X handle as a "cyber populist," said on X. "They wrote the legal predicate themselves and called it a brand." Girnus also noted that previous government efforts to limit the export of software, such as the attempt to put restrictions on powerful encryption techniques in the 1990s, have generally failed. Gary Marcus, a frequent critic of the AI industry, said in a social media post that he thought the government's action made sense, especially given its oft-stated position that the U.S. must stay ahead of China in the development of powerful AI systems. The national security directive would likely convince many Chinese-born AI researchers who currently work for labs such as Anthropic and OpenAI to return to China, he said. He added it would make investors question whether American AI companies were a safe bet, given the apparently capricious nature of the Trump administration's AI policy. Interestingly, Ball speculated that some people concerned that AI poses existential risks to humanity, including perhaps the AI safety-minded employees at Anthropic itself, might welcome the government's decision. This is because it might have the effect of slowing down AI development, something these AI safety proponents have been pushing for.

Anthropic lays out vision for how to bolster AI models' safety

Anthropic PBC co-founder Jack Clark said the company is in daily discussions with the Trump administration to resolve security concerns surrounding the startup's top two artificial intelligence models and outlined his vision for ensuring the safety of the emerging technology. While he declined to share specifics on talks with the government to ease U.S. export restrictions on its Fable 5 and Mythos 5 systems, Clark said that, like the administration, Anthropic and other companies are looking to address national security challenges raised by the rapid advances in AI tools. What emerges in terms of oversight will likely draw upon the know-your-customer standards governing the banking industry and also include tests of new AI systems, Clark said in an interview with Bloomberg's Odd Lots podcast. "That's the shape of, I think, where we will end up and what we're doing right now," said Clark, who now heads Anthropic's public benefit work. "It's initially going to be messy, but we're going to end up with a system on the other side." Anthropic has been negotiating with the administration since last week in an effort to reverse a Commerce Department directive requiring the company to obtain licenses for foreign nationals to use Fable 5 or Mythos 5. The unprecedented order from Commerce Secretary Howard Lutnick prompted Anthropic to disable access to both models pending a resolution. In a letter to Anthropic CEO Dario Amodei, Lutnick invoked U.S. laws that allow the government to impose export controls for national security reasons. Lutnick gave no basis for why the restrictions were necessary, but threatened the firm with criminal and civil penalties if it failed to comply, according to the letter, seen by Bloomberg News. So far, there's been little public sign of progress. When asked by reporters Wednesday during the Group of Seven summit in France about the status of talks with Anthropic, President Donald Trump said he thought they were "going fine" but didn't elaborate. Trump and Amodei took part in an AI-related event at the summit, but sat far apart during a lunch with tech executives and G-7 leaders. The dispute marks the latest clash between the administration and the company. Anthropic has been embroiled for months in a feud with the Pentagon over extra guardrails the company sought for military use of its AI tools. After contract talks broke down, the Defense Department declared the firm a supply-chain risk in March and sought other AI providers for the armed forces. For privately held Anthropic, the showdown comes at a delicate moment. The company filed paperwork to go public earlier in June and is expected to complete the process as early as October. That would follow SpaceX's record-shattering public debut this month and may coincide with one planned by rival OpenAI. Anthropic's and OpenAI's IPOs are expected to be some of the largest in recent history. Clark acknowledged public concerns surrounding the safety of AI and emphasized that the company tests its systems for potential failures. He reiterated the company's view, articulated in a recent blog post he co-authored, that there may need to be a system where governments and AI developers collectively decide when to halt work on the technology to minimize its risks. "The world needs an option to be able to potentially slow down or even in extreme circumstances, pause the development of this technology," he said. -- With assistance from Joe Weisenthal, Tracy Alloway, Josh Wingrove and Maggie Eastland.

Anthropic's own $13 billion investor helped get the White House to ban the world's most powerful AI model

A new report has shed light on the escalating situation between Anthropic, the US government, and the company's latest AI model, which was only available for 3 days before being forcibly taken offline by the Trump administration. Anthropic released Fable 5 and Claude Mythos 5 on Tuesday, which are the company's first "Mythos-class" models, seemingly classified as bleeding-edge AI. These Mythos-class models are straight-jacketed versions of the Mythos Preview that Anthropic showcased in April, and announced was too dangerous to be released to the public due to its power to identify cybersecurity vulnerabilities. Essentially, Fable 5 and Claude Mythos 5 are versions of Mythos Preview with safety rails designed to prevent the model from being used in nefarious ways, such as generating malicious code. These models with the built-in safety rails were banned by the US government over national security concerns, as authorities stated they had learned the Mythos-class models had been jailbroken. Reports state that Amazon, along with five other companies, contacted the White House and raised potential security concerns about Anthropic's Mythos-class models; as a result, the White House issued an export ban preventing non-US nationals from using them. Since that task was beyond Anthropic's capabilities, the AI company disabled the models for everyone. "Expanding on this success, today Amazon and Anthropic are deepening their collaboration with a commitment from Anthropic to spend more than $100 billion over the next ten years on AWS technologies. This encompasses current and future generations of Trainium (Amazon's custom silicon) and tens of millions of Graviton cores (Amazon's widely-adopted CPU chip) to provide superior price performance," reads an Amazon News press release from April, 2026 So, what do we take from this? A Wall Street Journal report claimed Amazon CEO Andy Jassy contacted US officials, including Treasury Secretary Scott Bessent, after researchers at Amazon used a series of prompts to get Fable 5 to generate malicious code. What's surprising is that Amazon is an investor in Anthropic, specifically $13 billion invested so far, since Anthropic's models run on Amazon Web Services infrastructure. Furthermore, Amazon has a contractual agreement with Anthropic that allows Amazon to consult with Anthropic management and raise concerns about significant business issues. So, Amazon, a major investor in Anthropic, stifled the release of Fable 5, which, if successful and driving more revenue, would have benefited both parties. These swift actions showcase the AI industry can be quite cut-throat, as you typically don't see business partners trying to squash the launch of their partner's best product. "Anthropic will secure up to 5 gigawatts (GW) of capacity to train and power their advanced AI models, including significant Trainium3 capacity expected to come online this year. The collaboration also includes a meaningful expansion of international inference in Asia and Europe to better serve Claude's growing international customer base," adds the press release Unless, of course, that business partner is making a competing product in that area, which Amazon is. In fact, Amazon has invested a similar amount in OpenAI, exceeding Anthropic's $13 billion by $2 billion, showing that AI companies are happy to invest in their competitors while simultaneously developing competing products. Amazon's Nova models are direct competitors to Anthropic's Claude models, and when it comes to the banning of the Mythos-class models, Anthropic states that the cybersecurity concerns raised by the five companies that led to the ban were replicable on OpenAI's latest model and other edge models.

Amazon Warning Triggered Anthropic AI Crackdown

Calls from Amazon CEO Andy Jassy and other tech firms prompted the Trump administration to suspend foreign access to Anthropic's Fable 5 and Mythos 5 AI models on Friday. The Trump administration's decision to cut foreign access to Anthropic's most powerful AI models was reportedly triggered by calls from Amazon CEO Andy Jassy. According to a report from The Wall Street Journal, Jassy contacted senior government officials on Thursday after Amazon researchers discovered a way to prompt Anthropic's Fable 5 model into returning information that could be used for cyberattacks. The call, along with warnings from at least five other firms, led to a frantic shuffle within the White House to gauge the threat and contact Anthropic CEO Dario Amodei, who reportedly pushed back on the administration's concerns and requests to voluntarily pull the model. "In reaction, the Admin issued the export control. The Admin did this reluctantly," said David Sacks, the co-chair of the President's Council of Advisors on Science and Technology, on Saturday. "It's been very surprised that Anthropic hasn't wanted to cooperate with a reasonable safety request (ie fixing the jailbreak issue)." Source: David Sacks The episode sheds light on what led to the US directive that forced Anthropic to pull its new model from the public on Friday night. Anthropic Claude is estimated to have around 18,900 monthly active users. In a blog post on Friday, Anthropic said it believed the US directive was the result of a misunderstanding about the threat posed by a "non-universal jailbreak," which came from an unnamed report. Amazon did not confirm if it spoke to government officials about Anthropic's models. "As a leading cloud provider that serves a large number of private and public sector customers, it's not uncommon for governments to seek our counsel on potential security risks," a spokesperson said. "When they occur, we don't share the details of these discussions." Anthropic said it is working to restore access for its users. "The Admin's hope now is that Anthropic remediates the safety issue," Sacks said, which would see the export control lifted, and Fable goes back into general release. "The Admin wants all of this to happen as soon as possible." AI tokens surge after Anthropic crackdown The episode has also demonstrated the US government's ability to promptly switch off access to US-based AI models on command, leading to a rally in decentralized AI tokens on Friday and Saturday. The native token of Bittensor, a decentralized AI protocol that lets people build and monetize artificial intelligence models, which some refer to as "the Bitcoin of AI," surged 23.9% over the past 24 hours. Venice Token (VVV), the native utility and privacy coin powering Venice AI, a decentralized, uncensored AI platform founded by Erik Voorhees, rose 16%. Near Protocol, a blockchain project building the infrastructure to support a decentralized AI agent economy, rose 6.2%.

Why Anthropic is halting access to its Fable 5 and Mythos 5 AI models

The US government issued a directive ordering Anthropic to suspend access to the models for foreign nationals. Anthropic has responded after the Trump administration issued a directive banning foreign nationals from using some of its most powerful AI tools, saying it would comply with the order but that it disagreed with the reasoning behind it. In a blog post late on Friday evening, the company said it had received a letter from the government at 5:21pm ET requiring it to suspend access to its Fable 5 and Mythos 5 AI models over national security concerns. The ban applies to foreign nationals both inside and outside the US and includes foreign national Anthropic employees, Anthropic said. "The net effect of this order is that we must abruptly disable Fable 5 and Mythos 5 for all our customers to ensure compliance," it continued. "Access to all other Anthropic models will not be affected". The company has apologised for the disruption to users and said it is working to restore access as soon as it can. Anthropic said its understanding was that authorities had become aware of a potential means of jailbreaking Fable 5. An AI jailbreak is when a hacker exploits vulnerabilities in an AI model, bypassing its ethical guidelines in an attempt to carry out restricted actions. The company sought to play down the issue, saying that the vulnerabilities appeared to be "relatively simple" and that it had found publicly available models were able to identify them. It added that while it was complying with the directive, it disagreed that a "narrow potential jailbreak should be cause for recalling a commercial model deployed to hundreds of millions of people". Anthropic also stressed that it had developed "strong safeguards that greatly reduce the likelihood that Fable is misused for tasks related to cybersecurity". Responding to the news on X, Jordan Bardella, a Member of the European Parliament and president of France's far-right National Rally party, said the move was a sharp reminder that artificial intelligence is a "major issue of national sovereignty". "Nations that do not quickly develop their own model(s) will always depend more and more on the choices of other powers: France must accelerate its support for the gem Mistral AI and the entire AI ecosystem," he added. It is not the first time Anthropic and the Trump administration have clashed. In February, Trump directed US federal agencies to immediately stop using Anthropic's technology after CEO Dario Amodei and his firm objected to its technology being used for certain defence purposes. "We don't need it, we don't want it, and will not do business with them again!," Trump wrote on Truth Social at the time, saying there would be a "phase out period" of six months. Anthropic later announced it would be taking legal action against the government after it also designated the company a "supply chain risk".

Anthropic sends staff to DC after model export restrictions

Senior technical staff from Anthropic are in Washington to meet with White House officials after the company pulled its latest AI models Friday in response to an order from the Trump administration. A source close to Anthropic confirmed to The Hill that top staff are in D.C. and have been meeting virtually with officials daily since the administration first reached out to the company. Anthropic disabled access to its new Fable and Mythos models Friday after the Trump administration issued an export control directive ordering the firm to block access to foreign nationals. The order cited national security concerns, which the company attributed to recent worries about a method of bypassing guardrails or "jailbreaking" its Fable model. It released Fable 5 and Mythos 5 last Tuesday. Fable 5, a "Mythos-class" model, was released to the general public with safeguards meant to block certain capabilities that the company considered potentially dangerous. Mythos 5, which has some of those guardrails removed, was released to a small group of cyberdefenders and infrastructure providers. That release marked a notable development, after Anthropic initially opted not to release Mythos more widely in the face of concerns that it could supercharge hacking capabilities. The company instead gave a small group of companies and government agencies early access to the model. Following the broader rollout of Fable and Mythos last week, Anthropic said it received "verbal evidence" from the U.S. government of a "potential narrow, non-universal jailbreak, which essentially consists of asking the model to read a specific codebase and fix any software flaws." In its statement Friday, the company sought to distinguish between universal and non-universal jailbreaks. It defined a universal jailbreak as the ability to broadly bypass a model's safeguards, suggesting no testers have discovered such a capability so far. Anthropic argued that perfect jailbreak resistance is not currently possible, adding it stands by its current defense strategy. "[W]e disagree that the finding of a narrow potential jailbreak should be cause for recalling a commercial model deployed to hundreds of millions of people," it wrote Friday. "If this standard was applied across the industry, we believe it would essentially halt all new model deployments for all frontier model providers." "As we have stated publicly, we believe the government should have the ability to block unsafe deployments, as part of a statutory process that is transparent, fair, clear, and grounded in technical facts," the company added. "This action does not adhere to those principles."

Anthropic and US Officials Meeting Monday to Resolve Dispute Over Export Curbs, Administration Official Says

WASHINGTON, June 15 (Reuters) - Senior ⁠Anthropic ⁠technical staff ⁠are scheduled to meet with government officials at the U.S. Department of Commerce in Washington on Monday, an official in the ⁠Trump ⁠administration said, after the U.S. government ordered the AI firm late last week to suspend access to its top-tier models for foreign nationals, citing national security concerns. Anthropic's technical ⁠staff ⁠have met with ⁠officials virtually every day since the Trump administration contacted the company ⁠on Friday, a person close to the company told Reuters. The Trump administration ordered Anthropic to block any ⁠foreign nationals, whether inside or outside the U.S., ⁠from using its latest models, Fable 5 and Mythos 5, the company said on Friday. In response, Anthropic said it would disable access to the models globally. The government believes there is a method of ⁠bypassing, or "jailbreaking," a safeguard that would prevent Fable 5 from being used in identifying software vulnerabilities, the company said. (Reporting by Karen Freifeld, Editing by Franklin Paul)

The Fable 5 AI Model Just Went Dark. What the US Government's Playbook Means for AI Companies -- and Every Business Leader

Fable 5, launched just three days earlier and already serving hundreds of millions of users, went offline within hours. The displacement of that commercial platform over a single narrow jailbreak claim is part of a longer and more revealing story. True Risk, Political Fight, or Both? In July 2025, Anthropic signed a $200 million Department of Defense contract to bring Claude into classified systems. Negotiations broke when the Pentagon demanded Claude be available for all lawful uses, including mass domestic surveillance and fully autonomous weapons. Anthropic refused, a decision made under significant political pressure that CEO Dario Amodei described as a matter of organizational conscience. The government designated Anthropic a supply chain risk, a classification never before applied to a US company. A federal judge found that action likely violated constitutional protections and issued a preliminary injunction. Friday's directive arrived through a different agency under different legal authority, but produced a similar result.

How Trump officials pushed Anthropic to shut down the world's most powerful AI models

That dramatic result came after a multiday back-and-forth involving numerous administration officials and input from tech industry leaders. The effects of the government's effectively shutting off access to one company's AI models could have profound implications for future AI policy. What follows is a timeline of the events that led to the effective ban on Anthropic's Mythos-class models. Thursday night (June 11) Thursday evening: Amazon raises concerns with Trump administration officials Amazon, Anthropic's biggest investor, contacted senior White House and administration officials, saying its researchers had identified what they believed was a way to coerce, or "jailbreak," the Fable 5 and Mythos 5 models into assisting users with risky cybersecurity questions. Anthropic released the models with safeguards designed to prevent responses to such queries. Axios reported that at least five other tech companies raised similar concerns with administration officials around the same time. Late Thursday night: Administration begins treating the issue as urgent White House and Commerce Department officials took Amazon's concerns seriously, worrying that foreign bad actors might bypass Anthropic's security guardrails and use the models to identify and exploit software vulnerabilities, Axios reported. The White House began trying to contact Anthropic, asking for an immediate reply.

Anthropic suspends access to Fable 5, Mythos 5, citing US directive

Anthropic has abruptly disabled its flagship AI models after a US government directive citing national security concerns. Anthropic has suspended access to its Fable 5 and Mythos 5 AI models after receiving a US government export control directive citing national security concerns. In a statement posted Friday, Anthropic said the government order requires it to suspend all access to Fable 5 and Mythos 5 by any foreign national, whether inside or outside the United States, including foreign national Anthropic employees. The AI firm said it received the directive at 5:21 pm ET, and the "net effect" of the order is to abruptly disable the models for all users to ensure compliance. It said the government did not provide specific details about the alleged threat, but said it believes authorities are concerned about a possible "jailbreak" method capable of bypassing the models' safeguards. However, Anthropic defended its model's guardrails, maintaining that over thousands of hours of penetration testing, a "universal jailbreak" -- a method to broadly bypass the model's safeguards -- had been found. "To date, the government has only given us verbal evidence of a potential narrow, non-universal jailbreak, which essentially consists of asking the model to read a specific codebase and fix any software flaws," said Anthropic.

Trump says negotiations with Anthropic are 'going fine'

US President Donald Trump stated that talks with artificial intelligence firm Anthropic are progressing well. This follows a meeting with Anthropic CEO Dario Amodei at the G7 summit. The administration has raised national security concerns regarding foreign access to Anthropic's advanced AI models. The company had previously blocked access to these models after a presidential order. US President Donald Trump said on Wednesday that negotiations with artificial intelligence company Anthropic are "going fine." Trump made the comment to reporters at the G7 summit after ⁠a ⁠meeting between leaders and tech bosses including Anthropic CEO Dario Amodei. Anthropic is in a dispute with the Trump administration over foreign access to its most advanced AI models, Fable 5 and ⁠Mythos 5. The company on Friday disabled access for all users to ⁠those models after Trump ordered Anthropic to block foreign nationals from accessing them. The meeting with G7 leaders and tech bosses was the first time that Trump has met with Amodei in public since the order. The Trump administration has said it has "national security concerns" with Anthropic's most advanced models. ⁠Spokespeople for Anthropic and Trump did not immediately respond to requests from Reuters to provide more detail about the discussions between the two.

U.S. Orders Anthropic To Halt Foreign Access To Its Most Advanced AI Models

The company said it received the export control directive on Friday from the government. June 12 (Reuters) - Anthropic said on Friday it has been ordered by the U.S. government to suspend access for all foreign nationals to its Fable 5 and Mythos 5 artificial intelligence models, citing national security concerns. The company said it received the export control directive on Friday from the government, which gave no specific details of its national security concern. It is Anthropic's understanding, however, that the government believes it has become aware of a method of bypassing, or "jailbreaking," Fable 5, according to the company's statement. "The net effect of this order is that we must abruptly disable Fable 5 and Mythos 5 for all our customers to ensure compliance. Access to all other Anthropic models will not be affected," Anthropic said. Anthropic added that it believed there was a "misunderstanding" and that it is working to restore access to the models as soon as possible. (Reporting by Ananya Palyekar in Bengaluru; Editing by Edmund Klamann)

Anthropic pulls plug on new AI models after Trump admin directive

Anthropic said Friday it will remove access to two AI models, Fable 5 and Mythos 5, to comply with a Trump administration directive restricting foreign nationals from using its latest systems due to security concerns. Details of the government's request, including the length of the restrictions, have not been made public. The company said it believes the order refers to a method to bypass its Fable 5 model. "We reviewed a demonstration of this specific technique being used to identify a small number of previously known, minor vulnerabilities," the company said in a statement. "These vulnerabilities all appear relatively simple, and we have found that other publicly-available models are able to discover them as well without requiring a bypass." The Hill has reached out to the White House and the Commerce Department for comment. Anthropic debuted a limited release of Mythos in April, sharing the powerful AI model with about 40 organizations that maintain critical computer infrastructure to patch security vulnerabilities. The company later expanded access to 150 organizations, and it unveiled Claude Fable 5 on Tuesday. This new model is a straightjacketed version of Mythos that includes additional guardrails to protect against cybersecurity threats. The White House has raised security concerns about powerful AI models, and President Trump signed an executive order earlier this month asking companies to voluntarily grant government oversight of new systems before making them public. Anthropic CEO Dario Amodei supported the president's directive, asserting in a statement on Wednesday that governments should have the power to block dangerous AI deployments that fail to meet safety standards. However, the company said Friday that the order does not meet the appropriate level of safety concerns. "As we have stated publicly, we believe the government should have the ability to block unsafe deployments, as part of a statutory process that is transparent, fair, clear, and grounded in technical facts," it wrote in the statement. "This action does not adhere to those principles." This is not the first time the AI firm has clashed with the federal government. Earlier this year, Defense Secretary Pete Hegseth labeled Anthropic a supply chain risk and prohibited the use of its Claude AI chat model following the company's dispute with the Pentagon over safeguards for this technology. The company sued the Trump administration over this designation and warned that the action "would set a dangerous precedent" for American companies negotiating with the federal government. Anthropic ended its Friday statement with an apology to its customers for "this disruption." "We believe this is a misunderstanding and are working to restore access as soon as possible," the company wrote.

Anthropic Says Has Taken Its Latest AI Models Offline to Comply With New Export Controls

WASHINGTON (AP) -- AI giant Anthropic said Friday it has taken its latest artificial intelligence models, known as Fable 5 and Mythos 5, offline to comply with a directive from the Trump administration to prevent their use by foreign nationals. The export controls mark the U.S. government's most significant step to date to restrict access to the most advanced AI models. Anthropic released Fable widely this week. That model is a limited version of the even more advanced Mythos, to which the company has tightly limited access due to cybersecurity fears. In a statement, Anthropic said it disagrees with the government's handling of the matter, saying it received the directive from the U.S. government Friday afternoon and it did not specify the national security concerns. "We believe the government should have the ability to block unsafe deployments, as part of a statutory process that is transparent, fair, clear, and grounded in technical facts," the company said. "This action does not adhere to those principles." Anthropic called it a "misunderstanding" and said it hopes to restore access to the models "as soon as possible." The Commerce Department did not immediately respond to a request for comment. The action comes 10 days after President Donald Trump signed an executive order to establish a framework for the federal government to vet the national security risks of the most advanced AI systems for up to a month before their public release. Participation by AI developers would be voluntary, the order said.

What Anthropic's Fable 5 Ban Reveals About AI National Security Risks

The banning of Fable 5, an advanced AI model developed by Anthropic, has brought to light critical issues surrounding the intersection of technology, security and regulation. As detailed by Wes Roth, the decision stemmed from a combination of factors, including the model's vulnerability to distillation attacks and a jailbreak flaw that bypassed safety protocols. These weaknesses not only exposed the system to misuse but also raised concerns about the broader implications of such technologies falling into the wrong hands. The involvement of the U.S. government, which cited national security risks and imposed export controls, further underscores the high stakes of managing advanced AI systems in a rapidly evolving landscape. In this feature, you'll gain insight into the specific vulnerabilities that led to Fable 5's downfall, including how distillation attacks and jailbreak exploits undermined its safeguards. Explore the role of corporate dynamics, such as Amazon's dual position as both investor and whistleblower and how these tensions influenced the regulatory response. Finally, understand the broader implications for AI governance, from the challenges of balancing innovation with accountability to the urgent need for clearer regulatory frameworks. This analysis offers a comprehensive look at the lessons emerging from this pivotal case. Security Vulnerabilities: The Core Issue The primary reason for banning Fable 5 lies in its significant security vulnerabilities, which posed serious risks to its safe deployment. Reports revealed that the model was highly susceptible to distillation attacks, a method where adversaries extract knowledge from an AI system to replicate or exploit its capabilities. Allegedly, some users transferred Fable 5's outputs to foreign developers, raising concerns about unauthorized access to sensitive AI technologies. Adding to the problem, researchers at Amazon identified a jailbreak vulnerability within Fable 5. This flaw allowed users to bypass the model's safety protocols, allowing it to perform unintended or potentially harmful tasks. While Anthropic initially implemented safeguards, such as "silent sabotage" mechanisms designed to limit misuse, these measures were later reversed due to public backlash over their restrictive nature. This reversal left the model exposed to exploitation, further amplifying security concerns and prompting calls for stricter oversight. Government Intervention and Export Restrictions The U.S. government played a decisive role in the banning of Fable 5, citing national security risks as a primary concern. Officials imposed export controls on the model, restricting its distribution to prevent the technology from falling into the hands of foreign adversaries. Reports of Chinese entities allegedly gaining indirect access to the model heightened the urgency of these measures, leading to swift government action. Anthropic's CEO, Daario Amade, participated in high-stakes discussions with government authorities to address these concerns. However, the feasibility of preventing universal AI jailbreaks and the adequacy of existing safeguards remained contentious topics. The lack of a clear regulatory framework for AI governance further complicated these discussions, leaving the government to adopt reactive measures in response to the growing risks. This incident highlights the pressing need for a more proactive and structured approach to AI regulation. Here are additional guides from our expansive article library that you may find useful on Fable 5. Corporate Dynamics: A Complex Web of Interests The controversy surrounding Fable 5 was further complicated by intricate corporate dynamics. Amazon, a major investor in Anthropic, also played the role of whistleblower by exposing the jailbreak vulnerability. This dual role has raised questions about Amazon's motivations, with some speculating that its communication with the White House may have influenced the government's decision to impose export controls on Fable 5. Anthropic's initial response to the security concerns was to implement "silent sabotage" safeguards, which subtly restricted the model's functionality to prevent misuse. However, these measures faced significant criticism from users and stakeholders, leading to their eventual removal. This tension between making sure security and maintaining user trust reflects a broader challenge within the AI industry: the difficulty of balancing innovation with accountability. The Fable 5 case illustrates how corporate interests, public perception and regulatory pressures can intersect in complex and often unpredictable ways. The Challenges of Accelerated AI Development The Fable 5 incident highlights the risks associated with the accelerated pace of AI development. AI labs face immense pressure to release new models quickly, often prioritizing innovation and market competitiveness over rigorous testing and security measures. This urgency can lead to critical oversights, as demonstrated by the vulnerabilities discovered in Fable 5. Moreover, the absence of established regulatory frameworks for AI governance leaves both corporations and governments struggling to address emerging risks effectively. Without clear guidelines, stakeholders are often forced to make reactive decisions that may not align with long-term security and ethical considerations. This reactive approach creates uncertainty and complicates efforts to manage advanced AI technologies responsibly. Implications for the Future of AI Governance The banning of Fable 5 raises urgent questions about the future of AI regulation and oversight. Enforcing export controls on AI models presents unique challenges, as digital technologies are inherently difficult to contain and can be disseminated across borders with ease. This incident underscores the need for stronger collaboration between AI developers, governments and international organizations to establish effective safeguards. At the same time, the tension between innovation and regulation continues to grow. While rapid advancements in AI drive progress and open new possibilities, they also introduce significant risks that demand careful management. Striking a balance between fostering innovation and making sure security will be critical as the industry evolves. The Fable 5 case serves as a stark reminder of the high stakes involved in developing and deploying innovative AI technologies. Addressing these challenges will require a coordinated effort from all stakeholders, emphasizing the importance of proactive governance and shared responsibility. Media Credit: Wes Roth Disclosure: Some of our articles include affiliate links. If you buy something through one of these links, Geeky Gadgets may earn an affiliate commission. Learn about our Disclosure Policy.

G7 leaders discuss 'trusted partners' access to cutting-edge U.S. AI models

EVIAN-LES-BAINS, France - Group of Seven leaders have discussed a plan to grant select "trusted partners" access to advanced artificial intelligence models from U.S. firms such as Anthropic, three diplomatic sources said on Tuesday, potentially opening a path around restrictions on non-American use. Anthropic on Friday disabled access for all users to Fable 5 and Mythos 5, its most advanced AI models. The company made that move after U.S. President Donald Trump ordered Anthropic to block foreign nationals from accessing its most advanced AI models, citing national security concerns. One of the diplomatic sources said a number of country representatives attending the ⁠annual summit of leaders of the G7 wealthy nations discussed the idea of widening access ⁠to advanced AI models with U.S. representatives. This was mainly with U.S. Commerce Secretary Howard Lutnick, on the sidelines of the opening G7 summit dinner on Monday in the ⁠French lakeside resort of Evian-les-Bains. The trusted partners could be countries or companies, said a second source, who declined to be ⁠named because the talks were ongoing. An agreement providing broader access to advanced models would allow G7 countries to use the models to develop stronger cybersecurity defenses against rivals such as China. A White House official said in a statement that the U.S. president's team has "an open line of communication with our allies, and we remain committed to addressing national security concerns with Anthropic's model." AI executives from Anthropic, OpenAI and Google, which are all developing highly advanced models, are expected to attend a working lunch on Wednesday to speak about technology issues, including regulation, AI infrastructure and networks. Anthropic's spokesperson did not immediately respond to requests for comment. Cybersecurity experts believe Anthropic's Mythos, a model designed to find flaws in computer code, may turbocharge attacks on banks' technology systems. The European Union is seeking access to Mythos ⁠to study the model's implications. Prior to Trump's order, Anthropic had given access to Mythos to select organizations in "more than 15 countries" so they could use the product to scan their computer systems for vulnerabilities, according to a company statement. The organizations included entities in the healthcare, communications, power and water sectors, according to the statement. The news of the trusted partners scheme was first reported by the Financial Times.

Sacks: US Wants To Lift Anthropic Export Controls 'As Soon as Possible' After Fable 5 Fix Sacks: US Wants

The US government plans to lift export controls on Anthropic PBC once the company addresses a "safety issue" that prompted the export suspension of its Fable 5 and Mythos 5 models. "The Admin's hope now is that Anthropic remediates the safety issue, the export control is lifted, and Fable goes back into general release," David Sacks, co-chair of the President's Council of Advisors on Science and Technology, wrote in a post on X. "The Admin wants all of this to happen as soon as possible." "We disagree that the finding of a narrow potential jailbreak should be cause for recalling a commercial model deployed to hundreds of millions of people," Anthropic said in its post. "If this standard was applied across the industry, we believe it would essentially halt all new model deployments for all frontier model providers." The move marks one of Washington's most forceful interventions in commercial AI to date. The dispute has quickly become a test of how aggressively the US will enforce safety standards on frontier models Anthropic 'At Odds' With Safe AI Anthropic has shut off access to both systems for all customers to ensure compliance. The administration, however, has argued that Anthropic demonstrated a "surprising" unwillingness to cooperate with its safety requests. "It's been very surprising that Anthropic hasn't wanted to cooperate with a reasonable safety request (i.e. fixing the jailbreak issue)," Sacks wrote. "Anthropic's reaction is very much at odds with their branding and ethos as a safe AI research community." Researchers at Amazon.com Inc. (AMZN.O) conducted jailbreak research that revealed vulnerabilities in Anthropic's model, according to the Wall Street Journal. The US government and Amazon were in contact about the vulnerability before the controls were imposed, according to people familiar with the matter. Amazon CEO Raises Concerns Amazon CEO Andy Jassy was among tech leaders who raised concerns to senior Trump administration officials this week about security risks in Anthropic's most advanced AI models, a person familiar with the matter told Reuters. "A highly credible trusted partner of both Anthropic and the USG who was testing Fable came forward with a jailbreak of those guardrails," Sacks said. He also criticized Anthropic for saying the jailbreak wasn't serious. "That is not what the trusted partner and the USG believe," he wrote. "Nor is that kind of minimizing language consistent with Anthropic's brand as the AI safety company." Some experts who favor export controls on advanced AI models found the Trump administration's action puzzling, Reuters reported. The decision affects US allies as well as adversaries. Jimmy Goodrich, a senior fellow at the University of California's Institute for Global Conflict and Cooperation, told Reuters the administration's decision wasn't "well thought-out." The ban prevents even Canadian and British employees at Anthropic from doing research and development, he said. Market News and Data brought to you by Benzinga APIs To add Benzinga News as your preferred source on Google, click here.

Anthropic disables top-tier AI models after US order limiting foreign access

Anthropic, an AI safety and research company, has dedicated teams to rigorously stress-test their models for potential risks and to design protective safeguards, but these defenses often remain fragile. Anthropic said on Friday it will "abruptly disable" its most advanced artificial intelligence models for all users after the US government ordered it to suspend access to the models for foreign nationals, citing national security concerns. The company received the export control directive to suspend access to Fable 5 and Mythos 5 for all foreign nationals, without being given specific details of its national security concern, Anthropic said in a statement. It is Anthropic's understanding that the government believes there is a method to bypass, or "jailbreaking," a safeguard that would prevent Fable 5 from being used to identify software vulnerabilities, the company said. The order comes just as a previous dispute between Trump administration officials and IPO-bound Anthropic showed signs of easing across parts of the US government. Anthropic's relationship with the government ruptured this year after it refused to allow the US military to use its AI models for domestic surveillance and fully autonomous weapons systems. The government responded by putting Anthropic on a supply chain blacklist, set to take effect later in the year. The action also marks a major escalation of US efforts to halt foreign adversaries' AI capabilities. For years, US export controls have focused on the chips and tools that power AI rather than on restricting foreign access to AI itself. Anthropic said the government has given it only "verbal evidence of a potential narrow, non-universal jailbreak". "We disagree that the finding of a narrow potential jailbreak should be cause for recalling a commercial model deployed to hundreds of millions of people," the company said. The government directive and Anthropic's response highlight growing tension between AI developers and regulators over how to assess risks from so-called "jailbreaks," or methods used to bypass model safeguards. As recently as Wednesday, Anthropic had called for greater US oversight of AI, including the ability to block models with unacceptable risks. It said, however, that the government action on Friday did not follow principles of fair and fact-based regulation. The Pentagon's chief information officer, Kirsten Davies, said in a post on X that the Defense Department supported prioritizing national security. "Some things are simply more important than revenue cycles, clickbait, and pre-IPO valuation. America First. Always," Davies said. Anthropic confidentially filed for a US IPO last month, edging ahead of rival OpenAI in the race to reach public markets. Mythos pushes US to regulate in AI development May reports by The Washington Post and The Wall Street Journal noted that Anthropic's Mythos model pushes the US toward more regulation in the AI development market. According to the WSJ report, US Vice President JD Vance was "alarmed" after a call with the heads of the biggest artificial intelligence companies, with the Mythos model among the most worrying because of its ability to find software vulnerabilities on its own. The main factor, according to the WSJ, is that these new models could target critical infrastructure administered by local authorities rather than the national government, with the local governments lacking the tools to disrupt such attacks when they occur. US National Economic Council Director Kevin Hassett said the Trump administration was working on a way to regulate how high-tech companies introduce new AI models to the market, with the main proposal being a system similar to the FDA's for testing new drugs. This would, according to Hassett, guarantee that "they're released to the wild after they've been proven safe," while an official working on the project told The Washington Post that the details of how it would work are "still being hashed out." Sophisticated cyberattacks Earlier this week, Anthropic rolled out an AI model named Claude Fable 5, representing a new tier of capability it calls "Mythos-class." The model is accompanied by guardrails barring its use in risky areas such as cybersecurity, which some users have complained are "overly broad," Anthropic said. Experts have said that Mythos models, in the wrong hands, could dramatically accelerate sophisticated cyberattacks, particularly in sectors such as banking that rely on complex, interconnected, and often decades-old technology systems. Anthropic said it had worked with the US government, among others, on safety ahead of the Fable launch and that models from rival AI providers showed a similar ability to unearth minor bugs in code. "The net effect of this order is that we must abruptly disable Fable 5 and Mythos 5 for all our customers to ensure compliance. Access to all other Anthropic models will not be affected," Anthropic said. Anthropic said that it believed there was a "misunderstanding" and that it is working to restore access to the models as soon as possible. "If this standard were applied across the industry, we believe it would essentially halt all new model deployments for all frontier model providers," the company said. Amazon's cloud unit AWS said late on Friday that Anthropic has asked it to revoke access to the models for "all users in all regions." A US official confirmed that the Commerce Department had issued an export control directive suspending all access to Fable 5 and Mythos 5 for foreign nationals. Dean Ball, a former White House official who contributed to the AI Action Plan the administration issued in the summer of 2025, said in a post on X that the order suggests all "non-Americans" would be restricted from using Anthropic's latest models, including those based in the US "This means you should expect to have to prove your citizenship to use Anthropic models," Ball said. Several key Anthropic personnel, including co-founder Chris Olah, AI researcher Andrej Karpathy, and philosopher Amanda Askell, were born outside the United States. Reuters was unable to determine their citizenship status, and an Anthropic spokesperson declined to comment on whether such staff would lose access to AI models.

Anthropic block marks US reversal, warning to Silicon Valley

America has taken a significant step by blocking foreign access to Anthropic's top AI models. This move follows concerns about bypassing security features in their latest AI. Officials are now discussing these security issues with Anthropic. This action highlights the US government's increasing focus on controlling advanced AI technology for national security reasons. The extraordinary move by the US to bar foreign access to Anthropic PBC's best AI models underscores the Trump administration's newfound willingness to exert control over a pivotal industry. It also reminds Silicon Valley that it's working with an imperfectly understood technology with uncertain impact. Washington has taken the unprecedented step of ordering the AI startup to disable access to its most advanced AI platforms for all foreign nationals. The US government issued the order after discovering it's possible to "jailbreak," or bypass the guardrails, of the Fable 5 AI model Anthropic released just days prior. The swiftness of the move triggered widespread concern over what the startup called a "disproportionate" response -- and a warning that such an approach could "halt all new model deployments for all frontier model providers." Top Anthropic personnel are now in talks with officials in Washington about the specific security concerns that've been raised, people familiar with the matter said, asking to remain unidentified to discuss private deliberations. One of the key officials in negotiations is Treasury Secretary Scott Bessent, who's sounded warnings for Wall Street about the potential dangers of frontier models such as Anthropic's Mythos: a platform that theoretically discovers and exploits flaws in software quicker than humans. The US government's response marked the most significant incursion by officials into an AI company's operations ever undertaken. It's unfolding just as the largest AI startups, including Anthropic -- already valued at more than $900 billion -- seek to become publicly traded companies. The surprise controls threaten to set a precedent for major AI model developers, including OpenAI, Alphabet Inc.'s Google and Meta Platforms Inc. The government is now willing to use extraordinary powers to compel AI developers to do its bidding on dealing with potential security threats -- despite a June executive order that declared it wouldn't force compliance on Silicon Valley. That earlier order declared the US wasn't creating a licensing regime for AI models, but the Trump administration has moved to do exactly that. "US frontier models are increasingly treated as strategic assets, with access tightly controlled and shaped by national security considerations," said Gary Tan, a portfolio manager at Allspring Global Investments. That's "a dynamic that is likely to persist as China continues to lag the US in compute." The debate is taking place against the backdrop of a race with China to dominate AI technology, as players like Alibaba Group Holding Ltd. and DeepSeek close the gap with their US rivals in performance and efficiency. Anthropic released Fable 5 last week as the first public-facing version of its Mythos-class model. The company has repeatedly warned about the cyber capabilities of Mythos, and White House officials have taken those concerns seriously, creating a pathway through the voluntary executive order for US agencies to use Mythos and other cyber-capable models to patch weak spots in their own systems. Washington's fears may extend beyond just the ability to jailbreak and influence any one platform. Broadly, the worry also is that adversaries may try to steal the AI model weights -- a valuable file of numbers that holds the key to the model's capabilities. The move to bar Fable and Mythos coincides with growing concern in Washington that AI companies must guard their trade secrets against possible insider threats. After decades of disseminating cutting-edge US technology around the world as a diplomatic and economic tool, Washington is now moving in the opposite direction. As with curbing the export of chips to geopolitical rivals such as China, the US is increasingly trying to preserve technology for home use, as a matter of national security. "This shows the US racing to use export controls to rein in frontier Al risks," said Stefanie Kam, an assistant professor at the China Programme, Institute of Defence and Strategic Studies at Singapore's Nanyang Technological University. "We can expect Washington to tighten curbs and treat Al exports as strategic leverage against Beijing. If the directive is narrowly targeted, US firms can adapt," she said. "If it's sweeping, it risks pushing innovation offshore while China advances." Last week, Senator Jim Banks, a Republican from Indiana who serves on a committee with export control oversight, called on Trump Administration officials to consider how the US could protect AI model weights from being stolen by China or other adversaries. Policy objectives aside, Silicon Valley is likely to resist Washington's attempts to exert control. From Meta to OpenAI, the biggest developers are racing to stake their claim on a technology that's likely to reshape whole economies and industries. Yet it's impossible to test every permutation of an AI model's potential capabilities, which leads to vague fears about threats and continuous scrutiny as the models keep improving. "There may be important capability thresholds -- such as AI systems that can automate large amounts of AI research and development in short periods of time -- that pose new kinds of risks," Banks wrote in a letter this month to officials including Bessent, National Cyber Director Sean Cairncross and Defense Secretary Pete Hegseth. "Industry experts have likewise raised serious concerns about the difficulty of keeping such systems under human control."

Anthropic Works to Strike Deal With White House on Mythos Ban | PYMNTS.com

As the Wall Street Journal (WSJ) reported Monday (June 15), the artificial intelligence (AI) startup sent top members of its tech team to Washington over the weekend in hopes of making a deal to end restrictions on two of its most powerful models. This came after the government banned foreign governments, businesses or individuals from using Anthropic's Mythos 5 and Fable 5 after learning that it was possible to "jailbreak" Fable 5's protections. The company shut down access to both models, while saying it disagreed with the administration's rationale. "If this standard was applied across the industry, we believe it would essentially halt all new model deployments for all frontier model providers," the company added. The WSJ, citing sources familiar with the matter, reported that officials from Anthropic and the administration held several hours of talks Saturday (June 13) on Fable 5, a pared-down version of Mythos designed for the general public. The report also noted that a group of cybersecurity experts and executives had published an open letter Sunday (June 14) calling on the White House to undo the ban. "This action has taken the best models away from defenders, created market uncertainty, and risked America's AI leadership without any real risk to justify it," the letter said. The letter also said it is "essential to provide AI to coders and security teams so they can find and fix flaws in their own newly-written as well as decades of legacy code faster than our adversaries." Meanwhile, a Monday report by Bloomberg called the ban "the most significant incursion" ever undertaken by a government into an AI company's business. It is happening as Anthropic and other AI startups are looking to go public, the report said, adding that it could set a precedent for other major AI developers, including Google, Meta and OpenAI. Anthriopic's troubles with the government did not begin with this ban. Earlier this year, the Defense Department declared the company a supply chain risk, meaning companies that want to work with the military would be blocked from also working with the AI startup. This came in the wake of objections from Anthropic over the military's use of its technology for mass domestic surveillance and autonomous weapons systems. The company has since sued the government to get the supply chain risk designation lifted. For all PYMNTS AI coverage, subscribe to the daily AI Newsletter.

Anthropic shuts down Mythos access after sweeping U.S. order

Anthropic has disabled access to its most advanced artificial intelligence models, including Mythos, following an unprecedented order by the Trump administration to keep the technology out of the hands of all foreign nationals. The U.S. government told Anthropic to suspend access to the Fable 5 and Mythos 5 models by any foreign national "whether inside or outside the United States," citing national security concerns, the company said in a statement. A U.S. official confirmed the Commerce Department sent the letter. The model developer has since shut off access to both systems to all customers to ensure compliance. Never before has the U.S. government taken such sweeping measures to rein in foreign access to frontier AI models developed by an American company. Both the Trump and Biden administrations have limited access abroad to other consequential technologies such as semiconductors and supercomputers, and some have debated the merits of blocking access to AI models. But restrictions on the software itself have raised constitutional and commercial concerns.

Amazon CEO Andy Jassy Triggered Ban On Anthropic's Mythos AI Models: Report - Amazon.com (NASDAQ:AMZN)

Jassy's Treasury Warning Sparks Global Shutdown The Wall Street Journal reported Saturday, citing sources, that Jassy told Treasury Secretary Scott Bessent that Amazon researchers used the Fable 5 model to obtain information usable in cyberattacks, prompting Anthropic's global shutdown of the models. On Friday, Anthropic disabled access to Fable 5 and Mythos 5 after the government directed the company to bar foreign nationals from the systems on national security grounds. The Information and Reuters similarly reported the concerns, attributing them to Amazon, a major Anthropic investor. Amazon did not immediately respond to Benzinga's request for comment. In April, Anthropic committed to spending over $100 billion on Amazon Web Services infrastructure, including Trainium chip capacity, while Amazon added $5 billion in fresh investment in the startup, building on its earlier $8 billion stake. Sacks Says Anthropic Refused Safety Fix David Sacks, former White House AI and Crypto Czar who currently co-chairs the President's Council of Advisors on Science and Technology, said on X that a "highly credible trusted partner" of both Anthropic and the U.S. government, testing Fable, found a jailbreak. Sacks wrote, "The Admin asked Dario to fix the jailbreak or de-deploy the model. Dario refused." Earlier, in a blog post, Anthropic said the capabilities apparently causing government concern are already available in other publicly accessible AI models. Photo courtesy: DFree On Shutterstock.com Disclaimer: This content was partially produced with the help of AI tools and was reviewed and published by Benzinga editors. Market News and Data brought to you by Benzinga APIs To add Benzinga News as your preferred source on Google, click here.

Anthropic's Weekend Drama Over Fable 5: Or Did Trump Sell Them a Lemon?

Or is it merely that the White House did not like the idea of a new AI model being released to the public without first getting clearance as Trump wanted When was the last time that the world's biggest IPO ever became history within a day of the stocks landing on the trading floor? Blame it on President Trump and his latest drama with AI startup Anthropic over the release of Fable 5. This safer version of their Claude Mythos AI model resulted in yet another run-in with the US government. As you'd have read in our reports, Anthropic released their most powerful AI model ever with built-in guardrails mid-week. And then followed it up with a draft policy framework on the future of AI that sounded like it came right out of the Trump playbook (does he have one is another question). But, none of it seemed to curry favour with the US administration. The Trump administration came out with a sudden export-control directive citing national security, that forced Anthropic to completely suspend access to Fable 5 and Mythos 5 models. The government claimed the ban was prompted by the discovery of a non-universal jailbreak technique that could allow the models to locate software flaws. Of course, the company complied with the directive but disagreed vehemently with the ban order. They argued that the identified vulnerability was narrow and could be replicated by any other publicly available AI models and that the commercial ban was unwarranted. At a broader level, the move also brought intense global debate surrounding AI sovereignty as well as the stability of using closed frontier models. However, the latest ka-pow between Anthropic and the Trump regime (there's a lawsuit over the White House terming them as a supply-chain risk) indicates that there could be more than what meets the eye going on here. For starters, why would a company formed with the single motive of building safer systems have a run-in over its latest model with guardrails? For starters, there is the moot question over the validity of the action where several former US officials and cybersecurity experts have raised concerns that it diverged from the hands-off approach to policing the AI industry that Trump had endorsed earlier this month. Of course, given his penchant for flip-flops, we can explain it away as merely another of those. On their part, Anthropic has clarified that a "perfect jailbreak resistance is not currently possible for any model provider" and that from its point of view the strategy of "defence in depth" is the best option that one can have. The company also vehemently disagrees the the issues raised around their latest model required to yank it off the shelves was. But that doesn't cut any slack with the likes of tech founder David Sacks who close to Trump and his gang. In a lengthy post on his X handle, Sacks says: "A highly credible trusted partner of both Anthropic and the USG who was testing Fable came forward with a jailbreak of those guardrails. The Admin asked Dario to fix the jailbreak or de-deploy the model. Dario refused." Some publications have reported that this highly trusted partner was none other than Any Jassy, CEO of Amazon. Ironically, it was Amazon that invested early in Dario Amodei and his team when they broke away from OpenAI over ethical issues. This multi-layered situation might take some time to fix is how we see things now. More importantly, there is also the question of timing. Defence Secretary Pete Hegseth once again took to his X handle to spread a bit of vitriol. "Three months ago, @DeptofWar kicked @AnthropicAI out of our building -- forever. Every passing day proves why that was the right move," he said. The announcement inflamed tensions between the Trump administration and Anthropic, which earlier this month confidentially filed for an initial public offering following a funding round valuing it at nearly $1 trillion. It also comes at a time when a few companies such as Microsoft have already held-off on using the model. Remember, Anthropic had introduced a new policy with Fable 5 to hold on to customer data for 30 days, which some haven't taken to kindly. Of course, the company claims it was part of a cyber defence strategy, but nobody seems to be buying into it. Maybe, Anthropic is growing too big for its boots. Having overtaken OpenAI and set considerable distance in terms of enterprise-level business growth, the company seems to have set its sights way higher than Sam Altman could have thought of, albeit due to their own work efficiencies as compared to the OpenAI boss's PR strategies. Maybe, it was the right time to clip their wings a bit. David Sacks has offered an olive branch. "The Admin's hope now is that Anthropic remediates the safety issue, the export control is lifted, and Fable goes back into general release. The Admin wants all of this to happen as soon as possible. It is frankly bewildered that Anthropic hasn't wanted to comply with safety requests that it previously said were its highest priority," he said in his X post. By the looks of it, even Anthropic does not want a long-drawn battle with the White House - not when the company is speeding towards making hay while the SpaceX Trojan Horse is still making news. And Sacks gave another peace offering: "The Admin values Anthropic's technical capabilities and feels that this issue, while serious, should be easily resolved. The ball is in Anthropic's court." However, a report by the NYT said Commerce secretary Howard Lutnick told Anthropic CEO Dario Amodei that companies would require a special licence to distribute the Mythos and Fable 5 models to "all destinations worldwide" and that failure to comply would result in criminal and civil penalties. Maybe, all of this is just another way for the White House to keep all AI companies under its thumb. Remember conversations around an executive order that asked technology companies to voluntarily let the government review their new models before releasing them to the public? Maybe, Amodie once again played the enfant terrible on this front and Trump just dished out a rap on his knuckles to get him back in line.

Ban on foreign access to Anthropic models highlights need for AI sovereignty - The Korea Times

Anthropic's ban of foreign nationals' access to its most advanced artificial intelligence (AI) models in response to a U.S. government order highlights how cutting-edge AI tools have effectively become strategic assets subject to government export controls, experts said Sunday. With the Korean government and firms restricted from accessing those models, the move has reinforced calls for Korea to hedge against such risks through a more strategic approach and by strengthening its AI sovereignty. Anthropic on Friday cut off foreign access to its Fable 5 and Mythos 5 models, saying the U.S. government believes there is a method of bypassing a safeguard against using the model to find cybersecurity holes. Mythos 5 is Anthropic's security-focused model designed to identify software vulnerabilities at an expert level, while Fable 5 is a safety-tuned version developed for broader public deployment. The latest restrictions were imposed in the form of a U.S. export control measure, according to Anthropic. News website Semafor reported that the move was prompted by concerns that a China-linked group had gained access to the model. Even before the latest restrictions, access to Mythos had been limited under Project Glasswing, a program that grants access only to vetted institutions due to concerns that the model's advanced capabilities could be used for malicious cybersecurity activities. Earlier this month, a small number of Korean organizations, including the Korea Internet & Security Agency, SK Telecom and Samsung Electronics, were granted access to the model through Project Glasswing. But the latest export control directive blocked their access as well. An official at the Ministry of Science and ICT said the government is "currently verifying the facts and discussing possible responses under the coordination of the Office of National Security." Experts said the restrictions demonstrate that advanced AI models are now being treated as strategic assets, much like semiconductors and advanced weapons systems, and are likely to face more stringent government oversight in the years ahead. "The restrictions show that Washington views advanced AI as a military-grade strategic asset from the perspective of economic and national security," said Lim Jong-in, chair professor at the Korea University School of Cybersecurity. "The message is that access will be granted only to trusted countries, and meeting that threshold will likely require extremely rigorous cybersecurity standards ... Korea should first focus on addressing the issue from an economic security perspective by leveraging the strategic advantage it already holds in semiconductor manufacturing," he added. "Ultimately, no advanced AI model can operate without semiconductors. Korea should capitalize on its strength in chip manufacturing to secure its position as a trusted partner and ensure continued access to cutting-edge AI technologies." As high-performance AI models such as Mythos have come to pose potential risks to critical infrastructure, ranging from power grids to financial networks, the landscape of economic security is now evolving. "The more powerful and influential an AI model becomes, the more likely it is to be subject to tighter export controls," an AI industry official said on condition of anonymity. "Just as semiconductors became a source of geopolitical leverage, countries that maintain technological leadership in AI will gain a strategic advantage. Companies will be forced to compete within that framework, making AI sovereignty all the more important." Ha Jung-woo, former senior presidential secretary for AI policy and future planning, wrote on Facebook, "There have been concerns that the most advanced AI systems could eventually become subject to export controls." He added, "As incidents like this can happen at any time, Korea should continue to pursue global cooperation while also securing its own capabilities to prepare for such scenarios." Korea is currently pursuing a national AI foundation model project aimed at fostering the development of domestic AI models. However, questions remain over whether the selected models will be able to compete with global leaders. Lim said replacing leading global models through the foundation model project will be an extremely difficult task, and the initiative should serve as a catalyst for nurturing the domestic AI industry and strengthening Korea's AI capabilities.

Anthropic staff to meet White House officials next week: Report

Senior Anthropic officials are in Washington to meet White House representatives. This meeting aims to resolve a dispute that has led to the shutdown of the company's advanced AI models. The models Fable 5 and Mythos 5 were affected by a Trump administration order. Anthropic responded by disabling global access to these models. Senior Anthropic technical staff are in Washington to meet with White House officials to try resolving a dispute that has taken the company's most advanced AI ⁠models offline, ⁠Axios reported on Sunday, citing a source close to the company. Reuters could not immediately verify the report. Anthropic and the White House did not immediately respond to requests for comment. Anthropic's technical staff have held virtual ⁠meetings with White House officials since the Trump administration's initial outreach on Friday, the ⁠report said. The Trump administration ordered Anthropic to block any foreign nationals, whether inside or outside the U.S., from using its latest models, Fable 5 and Mythos 5, the company said. In response, Anthropic said it would disable access to the models globally. The San Francisco-based AI startup, which has confidentially filed for a US initial ⁠public offering, had previously warned about the hacking capabilities of its Mythos model and held it back from wide release. Earlier this week, Anthropic rolled out a public version, called Fable, that included what it described as cybersecurity safeguards.

Anthropic Abruptly Pulls Fable 5 And Mythos 5 Days After Launch Following US National Security Order, Cal

Anthropic has abruptly disabled access to its flagship Fable 5 and Mythos 5 artificial intelligence models after the U.S. government ordered the startup to block foreign nationals from using the systems, citing national security authorities. The startup said it received the export-control order at 5:21 p.m. ET on Friday, and it covers "any foreign national, whether inside or outside the United States," including foreign-national employees. Anthropic responded by suspending access for all users, saying it was necessary to ensure compliance. Other models remain unaffected, the company said in a statement. Government Flags Potential AI Safeguard Bypass The company said authorities did not provide detailed justification for the directive. However, Anthropic said it understands the government's concerns relate to a potential method of "jailbreaking" Fable 5 -- a technique used to bypass a model's built-in safeguards. Anthropic pushed back against the decision, saying the reported technique revealed only "a small number of previously known, minor vulnerabilities" and did not demonstrate a broad failure of the model's safety protections. "We disagree that the finding of a narrow potential jailbreak should be cause for recalling a commercial model deployed to hundreds of millions of people," the company said. The directive lands amid an already fraught relationship between Anthropic and Washington. Earlier this year, the Pentagon designated the AI startup a "supply-chain risk" after it refused to relax restrictions on the military use of its models for autonomous weapons and domestic surveillance. Anthropic sued, alleging unlawful retaliation and violations of its free-speech rights, while multiple legal challenges remain ongoing. Why Fable 5 Matters The order also comes less than a week after Anthropic unveiled Fable 5 and Mythos 5, describing them as its most advanced AI systems to date. Fable 5 was notable because it brought capabilities derived from Anthropic's highly scrutinized Mythos program to a wider audience. Before launch, the company said the models underwent thousands of hours of testing involving government agencies, third-party researchers and internal teams. Anthropic maintained that no researcher has identified a "universal jailbreak" capable of broadly bypassing the models' safeguards and argued that the cybersecurity capabilities cited by officials are already available in competing frontier AI systems. The company said it is complying with the directive while working to restore access, calling the action a "misunderstanding" that is not "transparent, fair, clear, and grounded in technical facts." Image via Shutterstock Market News and Data brought to you by Benzinga APIs To add Benzinga News as your preferred source on Google, click here.

Exclusive | Anthropic floats proposal to commerce secretary to end US ban of powerful 'Mythos,' 'Fable' AI models: sources

Anthropic executives are pledging to work more closely with the White House in a proposal to Commerce Secretary Howard Lutnick as they scramble to resolve security concerns that triggered a crackdown on its powerful "Mythos" and "Fable" AI models, The Post has learned. The ongoing remediation talks have included a pledge by Anthropic to improve communication with the Trump administration and resolve any security concerns more quickly going forward, a source familiar with the situation said Thursday. Talks between Anthropic and Trump officials are progressing well, the source added, though an exact timetable for a permanent fix remains unclear. As of Thursday, Anthropic cofounder Tom Brown, the company's chief compute officer, and Sarah Heck, its head of public policy, are taking a leading role in negotiations with Trump administration officials, a second source with knowledge of the situation said. Representatives for Anthropic and the Commerce Department did not immediately return requests for comment. Anthropic first dispatched several top officials to Washington DC late last week after the Trump administration slapped export controls on "Mythos" and "Fable." The crackdown on Anthropic's models came after Amazon CEO Andy Jassy warned the administration that researchers had found evidence it was possible to bypass their safety guardrails. Anthropic opted to pull the models entirely, claiming it was the only way to comply with the export controls. The spat emerged as a key topic at the G7 Summit in France. President Trump said talks with Anthropic were "going fine" but didn't provide details, while Anthropic boss Dario Amodei urged world leaders to "resist the temptation to splinter" in their approaches to AI regulation. "Both parties are working quickly to get this resolved," an Anthropic spokesperson said earlier this week. "This is part of our ongoing commitment to working alongside the administration toward our shared goal of protecting US critical infrastructure and the US lead in cyber defense." Ironically, the White House's crackdown on Anthropic came just days after Amodei publicly argued that the US government should have the authority to shut down frontier AI models if it felt they were too dangerous. As The Post reported earlier this week, White House officials were irked that Anthropic and Amodei downplayed the safety flaw as a "narrow" problem, despite spending the last few years warning of potential AI catastrophe. A senior administration official said the White House had received warnings from "nearly half a dozen" companies besides Amazon before it took action. "Had Anthropic taken it seriously and, rather than dismissing it as isolated, moved to fix or pause access, this never would have happened," the US official said.

Anthropic Faces White House Heat as AI Curbs Threaten IPO Plans

Anthropic is working to ease tensions with Washington after restrictions on its top AI models raised regulatory concerns. The clash has sparked investor anxiety and could complicate the company's plans for a blockbuster public listing. Anthropic is racing to repair relations with Washington after the White House imposed restrictions on its most advanced AI models. This move has cast a shadow over the company's planned public listing and raised fresh questions about the future of frontier AI regulation in the United States. The company has launched an intensive outreach effort with senior administration officials after the government moved to restrict access to its flagship and Fable models over national security concerns. The restrictions came after reports that researchers had found ways to bypass some of the models' safety guardrails, prompting fears that the systems could be .

Anthropic cuts access to AI models over US 'national security' order - The Korea Times

SAN FRANCISCO -- Anthropic said Friday it has suspended access to two powerful AI models, Fable 5 and Mythos 5, to comply with a U.S. national security order. Just three days after publicly launching Fable 5, the company said in a blog post that it received a government directive banning all foreign nationals, even ones who work at Anthropic, from accessing Fable 5 and Mythos 5 over national security concerns. "The net effect of this order is that we must abruptly disable Fable 5 and Mythos 5 for all our customers to ensure compliance," it said. The company said it received the letter at 5:21 pm (2121 GMT) Friday. Axios reported that the letter came from U.S. Commerce Secretary Howard Lutnick. The U.S. Commerce officials did not immediately respond to a request from AFP. The firm said that the letter did not state what specifically concerned the government. However, the firm's "understanding is that the government believes it has become aware of a method of bypassing, or 'jailbreaking'" the Fable 5 model such that it could aid hacking. Fable 5, released Tuesday, is a locked-down version of Mythos 5, a cutting-edge AI model that Anthropic has held back from the public amid concerns that it had unprecedented abilities to identify software vulnerabilites -- or holes in code that hackers could exploit Mythos 5 -- the unrestricted model -- has only been released to select companies. The European Union, which gained access to Mythos earlier in June after weeks of talks, said the latest development further underlined "Europe's need for technological sovereignty." "We take note of Anthropic's statement and are assessing," said Thomas Regnier, a spokesman for the European Commission, which this month unveiled measures to slash the 27-nation bloc's dependence on America and Asia for key technologies, including AI. Anthropic said it had reviewed the "jailbreaking" method at the center of the speculation and the hacking opportunities it exposed, but it does not believe Fable 5 gives hackers capabilities that are not already available through other public models. The firm said that none of its security testers had found a "universal jailbreak," or a way to bypass it's safeguards against helping hackers. "We disagree that the finding of a narrow potential jailbreak should be cause for recalling a commercial model deployed to hundreds of millions of people," the company said. "If this standard was applied across the industry, we believe it would essentially halt all new model deployments for all frontier model providers." Anthropic has been locked in a legal standoff with the Trump administration for refusing to allow its technology to potentially be used for mass surveillance and autonomous weapons, leading the Pentagon to cut contracts with the company.

Anthropic executives to meet Trump officials to mend ties, CNBC reports By Investing.com

Investing.com -- Senior executives from Anthropic are meeting with Trump administration officials in Washington, D.C., on Monday to address the company's suspension of its latest artificial intelligence models, CNBC reported Monday, citing a source familiar with the matter. The meeting follows the AI company's receiving an export control directive on Friday that invoked national security authorities and ordered Anthropic to halt access to its newest AI models, Fable 5 and Mythos 5, for any foreign national inside or outside the United States. Anthropic disabled the models for all customers to comply with the directive. Access breaking news faster with institutional-grade feeds on InvestingPro -- get 55% off now. The action represents the latest tension between Anthropic and the government following an earlier conflict with the Department of Defense. The DOD designated Anthropic as a supply chain risk in March, which prohibited defense contractors from using the company's technology due to alleged national security threats. Defense Secretary Pete Hegseth commented on the directive in a post on X on Saturday, stating that each day demonstrates why blacklisting Anthropic was "the right move." The company launched Fable 5 and Mythos 5 on Tuesday, days before receiving the export control directive. Anthropic worked with government agencies to test the models before release and obtained approval to deploy them, the report said. Anthropic received no prior communication about a national security threat before the Friday directive, according to the report. Fable 5 and Mythos 5 expand on Anthropic's Claude Mythos Preview, a model designed to identify security vulnerabilities in software. The company restricted that rollout to selected companies as part of Project Glasswing, a cybersecurity initiative that appeared to gain approval from the Trump administration through several meetings about the model's capabilities. Anthropic described Fable 5 and Mythos 5 as advanced models that lead many industry benchmarks. Mythos 5 remains limited to selected users, while Fable 5 became available to enterprise customers and paid subscribers. The company said the wider release was enabled by new safeguards that block responses in high-risk areas, including cybersecurity and biology. In Friday's statement, Anthropic said it believes the government's concern involves a potential narrow jailbreak where a user could bypass a cybersecurity guardrail and ask Fable 5 to read a specific codebase and fix software flaws. "We disagree that the finding of a narrow potential jailbreak should be cause for recalling a commercial model deployed to hundreds of millions of people," Anthropic said. "If this standard was applied across the industry, we believe it would essentially halt all new model deployments for all frontier model providers." Anthropic characterized the dispute as a misunderstanding and said it is working to restore access as soon as possible.

Amazon voiced concerns about Anthropic AI models before US crackdown, source says​

The US government ordered AI firm Anthropic to halt its advanced AI models worldwide. This action followed concerns about potential misuse for cybersecurity vulnerabilities. Amazon CEO Andy Jassy reportedly raised security risks to Trump administration officials. Anthropic stated the government cited a method to bypass safeguards. The company complied with the national security order, disabling access globally. Amazon CEO Andy Jassy was among tech leaders who raised concerns to senior Trump administration officials this week about security risks in Anthropic's most advanced AI models, a person familiar with the matter told Reuters. Jassy's involvement sheds light on the extraordinary move by Anthropic on Friday to shut down its latest models globally in response to national security orders ⁠from President ⁠Donald Trump's administration. The San Francisco-based AI startup, which has confidentially filed for a U.S. initial public offering, had previously warned about the hacking capabilities of its Mythos model and held it back from wide release, but earlier this week, Anthropic rolled out a public version, called Fable, with what it described as cybersecurity safeguards. That brief release ended on Friday. In a blog post, Anthropic said the U.S. government told the company it believes there is a method of bypassing, or "jailbreaking," a safeguard against using the model to find cybersecurity holes. The bypass found only "minor" security flaws that other publicly available models can also find, Anthropic said in its blog post. The Trump administration ordered Anthropic ⁠to block any foreign nationals, whether inside or outside the U.S., from using both its latest models, Fable 5 and Mythos 5, the company said. In response, Anthropic said it would disable access to the models globally. Amazon did not confirm whether ⁠it spoke to government officials about Anthropic's models. "As a leading cloud provider that serves a large number of private and public sector customers, it's not uncommon for governments to seek our counsel on potential security risks," an Amazon spokesperson said. "When they occur, we don't share the details of these discussions." EXPORT CONTROLSThe Information, a technology news outlet, earlier on Saturday reported Jassy's concerns. The Information, citing a U.S. official, later reported that the administration was unlikely to force other AI firms to abide by restrictions similar to those placed on Anthropic. Reuters could not immediately verify the Trump administration's plans for regulating other firms. The U.S. government restrictions came in the form of an export control, Anthropic said in its blog post. The U.S. Commerce Department's Bureau of Industry and Security, which oversees export controls, did not immediately respond to a request for comment. Officials issued the export control "reluctantly" after Anthropic CEO Dario Amodei "refused" to "fix the jail break or de-deploy the model," White House adviser David Sacks wrote in a social ⁠media post on Saturday. "The Admin's hope now is that Anthropic remediates the safety issue, the export control is lifted, and Fable goes back into general release," wrote Sacks, co-chair of Trump's Council of Advisors on Science and Technology and formerly the White House's AI czar. Some experts who favor export controls on advanced AI models found the Trump administration's action puzzling because it affects allied nations as well as adversaries. "This was not well thought-out," said Jimmy Goodrich, a senior fellow at the University of California's Institute for Global Conflict and Cooperation. "It even bans Canadians and Brits employed at Anthropic from doing research and development." The order came just as a previous dispute between Trump administration officials and Anthropic showed signs of easing across parts of the U.S. government.

Trump admin open to talks with Anthropic over foreigner ban

The Trump administration is open to negotiating with Anthropic over its order barring foreigners from the company's most powerful artificial intelligence models, The Post has learned. A senior White House official, speaking on condition of anonymity to discuss the sensitive internal deliberations, suggested a resolution could be worked out directly with the AI giants CEO Dario Amodei. "We can't have frontier models running amok, the source briefed on the matter said. "It's all about working things through with Dario." The remark is the first indication the administration may view last week's export-control directive -- which forced Anthropic to disable its Fable 5 and Mythos 5 models worldwide -- as a starting point for talks rather than a permanent restriction.

G7 leaders discuss 'trusted partners' access for cutting-edge US AI models, sources say

EVIAN-LES-BAINS, France, June 16 (Reuters) - G7 leaders discussed a potential scheme to grant a limited number of "trusted partners" access to U.S. frontier models developed by AI giants like Anthropic and therefore win an exemption from a current ban on non-U.S. nationals, three diplomatic sources said on Tuesday. Last week, Washington decided to suspend access to Anthropic's most advanced AI models for foreign nationals, citing national security concerns. One of the sources said a number of delegates discussed the idea with U.S. representatives, mainly with U.S. Commerce Secretary Howard Lutnick, on the sidelines of the opening G7 summit dinner in the French lakeside resort of Evian-les-Bains. These "trusted partners" could be countries or companies, said a second source, who declined to be named because the talks were ongoing. A third source confirmed that no statement was expected on the matter on Wednesday, when tech issues will be on the G7 agenda. Cybersecurity experts believe Anthropic's Mythos, a model designed to find flaws in computer code, may turbocharge attacks on banks' technology systems, but it has so far not been made available to any European banks. The EU is seeking access to Mythos in order to study the model's implications. The news of the "trusted partners" scheme was first reported by the Financial Times. (Reporting by Michel Rose, Andreas Rinke and Julia Payne in Evian-les-Bains, FranceWriting by Gabriel StargardterEditing by Matthew Lewis) By Michel Rose, Andreas Rinke and Julia Payne

Top Anthropic staffers rush to DC in bid to reverse White House crackdown on 'Mythos' and 'Fable' AI models

Some of Anthropic's top staffers have rushed to Washington, DC, to try to reverse a White House crackdown on its "Mythos" and "Fable" AI models. Senior members of Anthropic's technical staff traveled to the nation's capital over the weekend to meet with White House officials, a person close to the company told The Post on Monday. Anthropic staff have held virtual meetings with the Trump administration since the White House slapped export controls on Mythos and Fable last week. Anthropic is committed to working with the Trump administration to resolve the dispute, the person said. A Trump administration official said Anthropic's team would meet in person with Commerce Department officials on Monday as part of those discussions. The White House implemented the export controls on Friday after receiving intel from Amazon CEO Andy Jassy that their guardrails could be bypassed in a way that threatened national security. Anthropic responded by pulling the models offline entirely, stating it was the only way to ensure compliance and that it disagreed with the White House's move. The White House referred a request for comment to the Commerce Department, which did not immediately respond. As The Post reported, White House officials were miffed that Anthropic downplayed the security flaw in Fable as a "narrow potential jailbreak" -- meaning security guardrails could be bypassed -- that was not unique to its models. That was despite frequent claims from CEO Dario Amodei and others that AI safety was their top priority. A senior administration official said the White House received warnings from "nearly half a dozen" companies -- not just Amazon -- that raised concerns about Fable before the government took action. Commerce Secretary Howard Lutnick and White House Cyber Director Sean Cairncross are among the officials that have engaged with Anthropic's team on the security concerns. Meanwhile, some notable cybersecurity officials have rallied to Anthropic's defense and called on the Trump administration to reverse course on the export controls. In a letter addressed to Lutnick and Cairncross, those officials said: "This action has taken the best models away from defenders, created market uncertainty, and risked America's AI leadership without any real risk to justify it." The more than 40 signatories on the letter included cybersecurity executives from major firms like Adobe and Zoom, as well as former Facebook chief security officer Alex Stamos. The experts argued that the "jailbreak" technique uncovered by Amazon and others is not a vulnerability unique to Fable and "can be replicated on GPT-5.5, Opus, Sonnet and even Chinese models like Kimi 2.7." Adam Thierer, a senior policy analyst at the think tank R Street Institute, called the Trump administration's action a "significant escalation in the politicization of AI and centralization of control over advanced computation in this country" in a post on X. The latest dispute with the White House is a major complication for Anthropic, which is already suing the administration over a move earlier this year by the Pentagon to label its AI models as a supply chain risk. Anthropic has been racing to go public ahead of rival OpenAI - and now has to contend with its major product being made unavailable to consumers. Amodei and other prominent tech CEOs are slated to be in Paris this week for the G7 Conference.

Anthropic staff to meet White House officials next week, Axios reports

June 14 (Reuters) - Senior Anthropic technical staff are in Washington to meet with White House officials to try resolving a dispute that has taken the company's most advanced AI models offline, Axios reported on Sunday, citing a source close to the company. Reuters could not immediately verify the report. Anthropic and the White House did not immediately respond to requests for comment. Anthropic's technical staff have held virtual meetings with White House officials since the Trump administration's initial outreach on Friday, the report said. The Trump administration ordered Anthropic to block any foreign nationals, whether inside or outside the U.S., from using its latest models, Fable 5 and Mythos 5, the company said. In response, Anthropic said it would disable access to the models globally. The San Francisco-based AI startup, which has confidentially filed for a U.S. initial public offering, had previously warned about the hacking capabilities of its Mythos model and held it back from wide release. Earlier this week, Anthropic rolled out a public version, called Fable, that included what it described as cybersecurity safeguards. (Reporting by Gnaneshwar Rajan in Bengaluru; Editing by Edmund Klamann)

Exclusive | Anthropic downplays security risks of 'Mythos' and 'Fable' AI models after ban -- prompting scorn from White House officials

Anthropic is downplaying risks that led the White House to ban foreign use of its powerful "Mythos" and "Fable" AI models - and US officials claim it's proof that CEO Dario Amodei has a one-sided, self-serving approach when it comes to cybersecurity. Late last week, Amazon CEO Andy Jassy reportedly alerted the Trump administration that its researchers had managed to "jailbreak" Fable - or bypass its safety guardrails - in a way that could threaten national security. The White House responded by slapping Mythos and Fable with export controls - restrictions that prevent the models from being used by foreign nationals and customers outside the US. Anthropic, in turn, responded by taking Mythos and Fable offline entirely. That's despite the fact that Anthropic claimed the flaw was a "narrow potential jailbreak" that was not unique to Fable, adding that it disagreed with the White House's handling of the matter. The company's seemingly contradictory words and actions drew scorn from Trump officials who noted Anthropic has spent months stoking fears about potential AI doomsday scenarios. "They seem obsessed with safety for everyone except themselves," a senior Trump administration official told The Post. "If I was an Anthropic investor, I would be extremely concerned." A second senior official with knowledge of the situation said the Trump administration had heard from "nearly half a dozen" companies that raised concerns about Fable before it took action - not just Amazon. The official declined to share which companies. "Had Anthropic taken it seriously and, rather than dismissing it as isolated, moved to fix or pause access, this never would have happened," the second US official said. Key administration leaders felt they couldn't trust Anthropic to manage their concerns - and those worries only grew after the call between Trump officials and Amodei on Friday, the Wall Street Journal reported. Company leaders reportedly pushed back on requests that they work with government experts. As recently as last week, Amodei published a controversial blog post declaring that the US government "should have the power to block" AI models that it deemed too dangerous. A source close to Anthropic claimed the Trump administration contacted key officials on Friday and gave them just 90 minutes to pull its models offline. The source also pushed back on reports that said Amodei was difficult to reach, asserting that both he and other senior staffers were available within 15 minutes of the White House's outreach. "There was no refusal to fix an issue. We were not presented with any details," the source said. Anthropic is eager to get its models back online and is working around the clock to address the administration's concerns, the source added. In a blog post, Anthropic claimed it had little choice but to take down the models in order to comply with the export controls and noted that some of its own employees are foreign nationals, meaning they would have been unable to work on the model. Treasury Secretary Scott Bessent and White House Cyber Director Sean Cairncross were among the officials who participated in frantic discussions with Anthropic leadership. White House AI adviser David Sacks, a vocal critic of Anthropic, said he was "frankly bewildered that Anthropic hasn't wanted to comply with safety requests that it previously said were its highest priority." "The Admin's hope now is that Anthropic remediates the safety issue, the export control is lifted, and Fable goes back into general release," Sacks wrote on X. The White House's decision was partly motivated by concerns that a China-linked group had surreptitiously accessed Mythos, Semafor reported. Anthropic already bars access to its products in China. Anthropic initially made Mythos available to about 40 top corporations, including Apple and JPMorgan, as part of a pilot program, warning it could cause a wave of devastating cyberattacks if given a wide release. Less than two months later, it released what it claimed to be the "safe" version of Mythos, dubbed Fable 5, for public use. "If you're Anthropic, what other outcome did you expect? You keep telling people your work is dangerous, you constantly talk in public about how you want the government to stop you, well, congratulations," said Perry Metzger, chairman of Alliance for the Future, a Washington, DC-based AI policy group. Since April, Anthropic executives also have publicly claimed that "Claude Mythos" AI model was too dangerous to be released to the public and argued it would be "good for the world to have the option to slow or temporarily pause" development. Amodei himself has a penchant for making terrifying predictions about AI - including that it has a 25% chance of destroying humanity. He has warned that AI could cause national unemployment to spike to 20% and said President Trump's decision to allow sale of advanced AI chips to China was like "selling nuclear weapons to North Korea." Ben Narasin, a general partner and founder of Tenacity Venture Capital, said Anthropic's dire warnings about AI had backfired. "He has become way too predictable with constantly crying wolf," Narasin said. "All of a sudden, you got what you asked for, but you didn't realize you didn't really want it." At the same time, Narasin said Anthropic may still emerge from this latest dustup as a winner. "It'll inconvenience Dario and it'll inconvenience OpenAI, but it'll massively hinder any new competitor that comes along," Narasin said. "So, he just built a new fortress wall protecting the castle." The warnings have sparked allegations from Sacks and others that Anthropic was using fear-mongering tactics as a method of "regulatory capture" - or crafting regulations in a way that benefits its interest at the expense of rivals. The dire warnings come, skeptics note, even as Anthropic has been scrambling to supercharge its growth. Earlier this month, the firm confidentially filed for an IPO ahead of its archrival OpenAI. Anthropic cofounder Daniela Amodei, Dario's sister, has said the company aims to tap the public markets for cash to wrangle massive amounts of computing power in data center deals with the likes of SpaceX and Google.

US blocks foreign access to Anthropic's most advanced AI models, Axios reports

June 12 (Reuters) - The Trump administration is blocking foreign governments, companies and individuals from accessing Anthropic's most advanced AI models, Axios reported on Friday. Reuters could not immediately verify the report. U.S. Commerce Secretary Howard Lutnick on Friday sent a letter to Anthropic CEO Dario Amodei saying that the Mythos 5 and Fable 5 models would be subject to export controls to any location outside of the U.S. and to all foreign persons within the country, the report said. (Reporting by Ananya Palyekar in Bengaluru; Editing by Edmund Klamann)

Anthropic and US govt in talks to create AI security framework after Claude Fable 5 ban: Report

ust a few days after launch, Anthropic removed access to both models globally, citing an export control order from the US government. Anthropic and the US government are reportedly working together on a new framework to assess security risks in AI models. The move comes after the government imposed export controls on Anthropic's recently launched Claude Fable 5 and Mythos 5 models over concerns about a security vulnerability known as a jailbreak. According to a report by Politico, officials from the White House and Anthropic are discussing a common set of standards that could help determine how serious future AI security flaws are and what kind of government action may be needed. For those unfamiliar, Anthropic launched Claude Fable 5, the first publicly available Mythos-class AI model, last week along with the more advanced Mythos 5 model. Just a few days after launch, Anthropic removed access to both models globally, citing an export control order from the US government. Also read: OpenAI improves health intelligence in ChatGPT, here is how Now, both sides are trying to create a clearer process for handling similar situations in the future. The discussions are being led by Anthropic's Head of Public Policy Sarah Heck and co-founder Tom Brown. The report says negotiations have improved in recent days. Earlier, discussions had nearly broken down after Anthropic rejected demands to remove Fable from the market. However, over the weekend, senior administration officials held multiple calls with Anthropic leaders, including Tom Brown. The conversations were followed by several in-person meetings in Washington. Anthropic reportedly sent senior researchers and safety experts to meet officials at the Commerce Department earlier this week. Also read: Anthropic CEO Dario Amodei takes aim at OpenAI, says we'll see who wins Meanwhile, Anthropic is also facing new legal trouble as more than 100 authors accused it of using their books without permission to train its AI models. The lawsuit has once again raised concerns about how AI companies collect and use copyrighted content. The case was filed in the US District Court for the Northern District of California by a group of writers led by Thomas William Shakespeare. The authors claim that Anthropic used hundreds of copyrighted books without obtaining proper licences or permission from the creators. According to the complaint, the company used around 500 copyrighted works as training material for its AI systems.

Anthropic wants Claude Fable 5 and Mythos 5 back online, sends execs to US for talks

Now, the AI company is reportedly trying to restore access to the models. Anthropic launched Claude Fable 5, the first Mythos-class AI model designed for public use, last week. Alongside it, the company also introduced Mythos 5, a more advanced version of its restricted AI model. However, just a few days after launch, Anthropic removed access to both models globally, citing an export control order from the US government. Now, the AI company is reportedly trying to restore access to the models. According to a report by Axios, Anthropic has sent senior executives to Washington to meet with US officials. The discussions will likely aim to resolve concerns that led to the restrictions. US authorities are reportedly concerned about the capabilities of the new AI models and whether the safeguards built into them are strong enough to prevent misuse. The government is also worried that users outside the US could gain access to AI tools that might be used for cybersecurity-related activities. That's why Anthropic receieved an export control directive to restrict access to the models. The restrictions affect users around the world, including those in India. Also read: Anthropic removes access to Claude Fable 5 and Mythos 5, here is why According to the report, Anthropic's senior technical staff will meet with officials to explain the safety measures included in the models. The company is also expected to discuss possible changes that could address the government's concerns and potentially allow the models to become available again. Despite following the government order, Anthropic has maintained that it disagrees with the decision. "We disagree that the finding of a narrow potential jailbreak should be cause for recalling a commercial model deployed to hundreds of millions of people. If this standard was applied across the industry, we believe it would essentially halt all new model deployments for all frontier model providers," the company said in a blogpost last week. "We apologise for this disruption to our customers. We believe this is a misunderstanding and are working to restore access as soon as possible." Also read: OpenAI sued after ChatGPT allegedly urged woman toward suicide, complaint claims The restriction has also sparked discussions about dependence on foreign AI technologies. Zoho co-founder Sridhar Vembu said the situation should serve as a warning for countries that rely heavily on foreign technology platforms. He urged governments and businesses to invest in domestic AI development and open-source alternatives.

Amazon triggered the White House ban on Anthropic Fable, here is how

The move has sparked fresh debate over AI safety and regulation. Anthropic this week launched its much-anticipated Claude Fables, a safety-focused model built on Claude Mythos. However, in a blog post published on Friday, the company said it had been instructed by the US government to revoke access to its most advanced AI models for certain foreign users and to suspend access more broadly. Reports indicate that the decision followed a call from a technology industry executive to senior US officials, triggering a government crackdown on Anthropic's Claude Mythos and Claude Fable models. The current crackdown from the government has shown increasing conflict between AI, national security, and politics. It has also raised questions about who should decide when an advanced AI system is too dangerous to share, as governments and tech companies race to stay ahead in this fast-changing field. According to a report by The Wall Street Journal, Amazon CEO Andy Jassy told administration officials that company researchers had managed to get Anthropic's Fable 5 model to provide information that could potentially aid cyberattacks through a series of prompts. The findings were shared with officials including Treasury Secretary Scott Bessent, prompting discussions inside the White House about possible responses. Also read: Anthropic removes access to Claude Fable 5 and Mythos 5, here is why The report said security researchers later examined Amazon's claims, and officials concluded that restricting access for foreign governments, companies and individuals was the quickest way to address the perceived risk. President Donald Trump reportedly approved the action despite concerns that such a move could hurt American competitiveness in AI. Anthropic has pushed back against the government's assessment. The company argues that the information highlighted by Amazon involves relatively basic software vulnerabilities that can also be identified using other publicly available AI models. It also disputes suggestions that the issue represented a complete bypass of the model's safety protections. Some cybersecurity experts familiar with the research have expressed similar views. They say the reported findings did not demonstrate access to the most sensitive capabilities of the model, including tools that could generate working exploit code for attacks. Also read: Switched to Claude from ChatGPT? Fix these 3 privacy settings right now The situation has also revived long-running tensions between Anthropic and the Trump administration. The company has previously clashed with government officials over AI policy and military use of its technology. Critics of the latest decision argue that politics may have played a role alongside security concerns. Rumours about Anthropic exploring a public listing in the coming months have been going around for a while. However, the restrictions on the AI models could have significant business consequences for the company, as customers can weigh alternative AI providers like OpenAI or Google.

Anthropic removes access to Claude Fable 5 and Mythos 5, here is why

Anthropic has suddenly removed access to both AI models for all users. Anthropic launched Claude Fable 5, the first Mythos-class AI model designed for general use, earlier this week. Alongside it, the company also introduced Claude Mythos 5, an updated version of its restricted AI model. Now, just a few days after launch, Anthropic has suddenly removed access to both AI models for all users. The decision follows a legal order from the US government, citing national security authorities. In a blogpost, Anthropic said it received an export control directive from the government. The order required the company to stop access to Fable 5 and Mythos 5 for all foreign nationals, both inside and outside the US. "The net effect of this order is that we must abruptly disable Fable 5 and Mythos 5 for all our customers to ensure compliance," the AI company said. The letter offered no details about the alleged national security concern. However, Anthropic believes the government may have found a technique that bypasses or jailbreaks Fable 5. Also read: OpenAI sued after ChatGPT allegedly urged woman toward suicide, complaint claims Anthropic further stated that, so far, the government has offered only verbal evidence of a narrow, non-universal jailbreak that appears to involve asking the model to analyse a specific codebase and fix software flaws. Anthropic explained that it reviewed the report and found that the results were not unique to Fable 5 and that other publicly available AI models (including OpenAI GPT-5.5) can already do similar tasks. "We have not even received a disclosure of a concerning non-universal potential jailbreak that led to a harmful result. The potential jailbreaks that have been disclosed to us are either entirely benign responses or are minor findings that provide no Mythos-specific uplift," Anthropic said. Anthropic also defended the safety systems built into Fable 5. Before launch, the company worked with government agencies, outside security researchers, and internal teams to test the model for possible misuse. According to the company, no testers found a broad method that could fully bypass the model's protections. Anthropic also admitted that no AI model today is likely to be perfectly protected against jailbreak attempts. Also read: AI may bring cyberattacks and job losses if left unchecked, warns Anthropic CEO Dario Amodei Despite following the government order, Anthropic said it disagrees with the decision. "We disagree that the finding of a narrow potential jailbreak should be cause for recalling a commercial model deployed to hundreds of millions of people. If this standard was applied across the industry, we believe it would essentially halt all new model deployments for all frontier model providers," the company said. "We apologise for this disruption to our customers. We believe this is a misunderstanding and are working to restore access as soon as possible."