Five Eyes Alliance Warns AI Cyber Risks Will Transform Hacking in Months, Not Years

I Met With China's Top AI Experts. They're Freaking Out, Too

Just over a week ago, I attended a major artificial intelligence conference in Zhongguancun, Beijing's bustling high-tech district. It was packed with fascinating sessions touching on everything from recursive self-improvement -- the idea that models can tweak their own code and advance indefinitely -- to humanoid robots. And it featured a few legends of computing, including Whitfield Diffie, co-inventor of public-key cryptography, and Andrew Barto, who won the Turing Award with Rich Sutton for his pioneering work on reinforcement learning. But I left with one takeaway above all else: The US and China should put their fierce AI rivalry to the side. Frontier AI's cybersecurity and systemic risks are too serious to ignore, and increasingly capable agentic models could soon cause chaos unless the world's AI superpowers can work together. "AI is a global technology with global benefits, global harms, and a consistent tendency for new capabilities to eventually proliferate," Stephen Casper, a computer scientist at MIT who spoke at the conference via video, told me afterward. Until now, the US has largely viewed China's AI advances as an economic and national security threat. Washington has imposed tight restrictions on chips and chipmaking equipment to stymie the country's development of powerful AI. Most recently, the US government ordered Anthropic to prevent foreign nationals from accessing its most powerful models, Mythos and Fable 5, over national security concerns. In response, Anthropic revoked access for everyone. One company that was of particular concern, WIRED previously revealed, was a South Korean telecom giant with alleged ties to China. But the conference, organized by the Beijing Academy of Artificial Intelligence, reinforced the idea that both the US and China stand to lose if AI is developed too quickly and recklessly. As AI becomes more powerful, more agentic, and more intertwined with everyday life, the risks that it could be used to conduct cyberattacks or fail in catastrophic ways will only grow. Because the world's two dominant AI powers are responsible for the most advanced models, cooperation between them feels like it will be crucial. Casper pointed to research showing that the benefits of international collaboration on AI dangers outweigh any national security risks that come from working together. He likened the current situation to how the US and the Soviet Union were forced to work together on nuclear dangers, even as they sought to out-stockpile one another. "One thing that almost everyone in AI can agree on right now is that AI doesn't need a Chernobyl moment," Casper said. One day-long session highlighted the universality of the cyber challenges raised by more advanced AI. This includes new kinds of vulnerabilities in AI-generated code, novel ways of attacking systems enabled by agentic tool use, and automated methods for carrying out social engineering attacks. After another session, I spoke with Lin Yun, a professor at Shanghai Jia Tong University who does excellent work on AI and computer security. Yun told me he expects hackers to gain an advantage over the near term, but that new countermeasures, including novel uses of AI, should tip the balance back toward defense over time. Yun said that even if international cooperation is complicated by competition, it should remain a priority. "If different countries understand the risks in similar ways, it becomes easier to develop shared safety principles and technical standards," he told me. "The key is to find areas where sharing can reduce systemic risk without exposing sensitive operational details." Perhaps the most pressing question for both nations is how to balance openness with risk. Open-weight models have become crucial for research and innovation, with Chinese models proving popular in the US. But as these models advance, it will become more challenging to ensure they don't help hackers identify security vulnerabilities and can't be wielded as cyber weapons. Over the last few years, Chinese companies have taken the lead in offering highly capable open-weight AI models, like Moonshot's Kimi, Alibaba's Qwen, and Z.ai's GLM. The US has rebooted its own open-weight AI push with models like Nvidia's Nemotron. But we're approaching an inflection point where even less powerful open models could prove dangerous if they're stripped of guardrails. The latest model from China's Z.ai, GLM 5.2, includes frontier agentic and coding capabilities, according to expert analysis. The next generation of open-weight AI models might be just as capable as Fable or Mythos. In fact, just this week, 360 Security Technologies, a top Chinese cybersecurity firm, said it had developed an AI model with hacking capabilities on par with Mythos. Yun said the industry will need to devise new ways of guaranteeing that open models are up-to-date, free of backdoors and vulnerabilities, and have met safety standards. One possible sign of things to come? A source at one of China's leading AI companies, who asked to remain anonymous because they weren't authorized to talk to the press, tells me that security concerns are one reason why some advanced models in China are no longer being released as open source. This is an edition of Will Knight's AI Lab newsletter. Read previous newsletters here.

How big a cybersecurity threat are the latest AI models, really?

We spoke to an expert about what the risks are and how we can protect ourselves Artificial intelligence is getting better at everything, including hacking. It's becoming easier than ever before to steal someone's identity, cripple sensitive banking and health care systems, or hold a company's data ransom. And if cybersecurity defenders aren't ready, cyber attackers will exploit AI to wreak havoc. "The timeline is not years, it is months," the multinational intelligence group Five Eyes warned June 22. The newest AI technology "lowers barriers for malicious actors and increases the speed and complexity of attacks." Five Eyes is a secretive alliance dating back to World War II in which Australia, Canada, New Zealand, the United Kingdom and the United States work together to gather intelligence or respond to security threats. Two new models, Anthropic's Mythos 5 and OpenAI's GPT-5.5, have each proven capable of independently planning and carrying out a full takeover of a simulated corporate network. That means a single hacker could do what once required a large team, says AI security expert Michael Alexander Riegler of Simula Research Laboratory in Oslo, Norway. These models can also find and exploit security holes in operating systems, browsers and other software at an expert level, which could leave defenders scrambling to patch vulnerabilities. The Five Eyes warning comes on the heels of the U.S. government barring Anthropic from allowing foreign nationals access to Mythos 5 and another new model, Fable 5, citing national security concerns. Mythos 5 had been made available only for cyber defenders to help identify and fix any vulnerabilities before the tech landed in the hands of bad actors. Fable 5, a version of the same model loaded with extra safeguards geared toward preventing its misuse in cybercrime, was available to the general public for only a few days. So are AI-fueled cyberattacks really an imminent threat? Or is this more corporate posturing and marketing hype? Science News asked Riegler about the risks and the reality. This interview has been edited for length and clarity. SN: Are the latest AI models especially dangerous? Riegler: In the last months, we heard a lot about Mythos and how dangerous it is. And I agree that AI has a lot of security risks. When the capability goes up for these models, the time from finding any issue to exploiting it gets really short, because you can basically automate the whole pipeline. But it's not something really new.... [It's] not just the latest models [that] are a security threat, but also other models that are already available. If you know how to use them, you can ... do quite bad stuff. It's logical if you think about it. Tools like Claude Code make it much, much more efficient to code. You can automate the process. You could use several hundred [AI] agents at the same time to explore different security holes. Before, you needed to hire a group of two to three hundred hackers [for organized cybercrime]. Now you maybe just have to buy 300 GPUs [specialized computer chips used to run AI] and you can do similar things. SN: So why all the concern about Mythos? Riegler: I think it's as much marketing as a real danger. If you say, "I'm sitting on something that is so dangerous, we cannot release it," a lot of people will get really interested in that and want to be part of this group that has access.... It's a bit of a show, and [the U.S. government and Anthropic] are focusing on the wrong problem. SN: What is the right problem to focus on? Riegler: AI is a huge risk for security.... But [the security risk] is not just about the model. It's also about everything around the model. What kind of tools you provide it, if it has access to internet, if it can test its own code. So the whole system around it is also very important. In our tests [with systems combining small AI models and various tools], we made a system that could, for example, hack your website and find security holes in your website, but also hack your network and try to find security holes there. Or it could break another AI and get it to do things it shouldn't do. It's quite flexible. SN: Is there an upside to the fact that cybersecurity defenders will have access to the same tools as attackers? Riegler: The testing of the security of your own system will be more efficient. I think, in the end, it will balance itself out. It will be again this cat-and-mouse game of who finds the hole first, who closes it first, or who exploits it first. Just at a much higher speed than we see now. SN: What can people do to protect themselves from sophisticated AI-enabled cyberattacks? Riegler: Be even more careful about using different passwords for different services. Have your software up to date all the time, use two-factor authentication. Everything you do that is maybe a bit bothersome, but increases security, I would recommend you to do. SN: What about companies and public agencies? Riegler: When I talk to security experts in different companies or the public sector, they're still behind. Some of them are very scared, others are not at all. They have to take AI security risks seriously and not think that it's something far in the future.

Spy agencies say AI can help combat AI cyber risks. But don't forget the basics

Cybersecurity agencies of Australia, Canada, New Zealand, the United Kingdom and the United States issued a call to action on Monday for cyber defenders. The message was clear: artificial intelligence (AI) is a powerful weapon for cyber attackers; defenders must act urgently to improve their cyber defences. There is much hype and uncertainty surrounding AI and cybersecurity right now. This latest statement comes little over a week since the US government caused frontier AI provider Anthropic to block access to Mythos and Fable, its most advanced AI technology, over fears they might be misused by foreign adversaries to attack US government systems. In this torrid environment, it's important for cyber defenders to look past the noise and prioritise what is truly important in protecting their systems. A call to arms The joint statement was issued by the heads of the national cybersecurity agencies of the Five Eyes. It warns that AI is dramatically shifting cyber risk and spells out how defenders must act to secure their organisations. It notes how powerful AI is already helping adversaries carry out more sophisticated attacks more quickly. One way this is happening is through automated vulnerability discovery and exploitation. No software is perfect. Adversaries leverage subtle design or implementation flaws in a system's software to break into that system. They then take control of it and use it as a staging ground to launch further attacks. This is why it's so important for cyber defenders to keep up to date with deploying software patches. These are small modifications to system software that close off known vulnerabilities. AI is enabling adversaries to find flaws orders of magnitude faster, as well as to work out how to exploit those flaws to carry out attacks. For this reason, the Five Eyes statement warns that AI is dramatically shrinking the time between when a vulnerability is first discovered and when it is first exploited in an attack. Defenders can no longer afford to wait weeks before deploying software patches. What can defenders do? The Five Eyes report notes cyber fundamentals are crucial and encourages organisations to use AI to boost defences. But deploying AI without first investing in cybersecurity basics would be a mistake. The cyber defenders who will be able to weather the AI storm will be those who already have mature practices. They know exactly what assets they need to protect, which systems in their organisation are exposed to attack, and what defences are in place to protect exposed systems. They also know to measure defence effectiveness and determine where defences are missing. They also use evidence-based processes for tracking known vulnerabilities in their systems and prioritising which are most important to patch. These are backed up by reliable processes for rapidly testing and rolling out software patches, as well as for responding to cyber breaches and incidents. When AI makes finding software vulnerabilities cheap, the next generation of software needs to be engineered to be secure by construction. Working out the best methods to do this is what I have devoted my research career to. Before reaching for AI, defenders should first invest in their fundamentals. Otherwise, they are effectively deploying a robot guard dog to defend an unlocked door. The role for AI in cyber defence This doesn't mean AI can't play an important role for cyber defence - just that it should augment rather than replace strong cyber fundamentals. AI benefits attackers and defenders alike. An AI model that can help attackers find software vulnerabilities can also help defenders fix those same vulnerabilities. AI that can automatically exploit software vulnerabilities is just as useful to defenders in helping them to confirm their software has been correctly patched. AI that can map and discover sensitive assets within a computer network is useful for both offensive and defensive purposes. This is why it's so important that defenders have access to AI capabilities, so they can be leveraged to harden and protect systems before that same AI is used to attack them. Can regulation help? Working out how to balance the competing benefits and risks of new cybersecurity technology is nothing new. In the 1990s, society grappled with how to regulate the encryption that protects online communication from adversaries but also allows them to avoid law enforcement. In the 2000s the rise of cyber exploit kits allowed defenders to better test their systems but also enabled any disaffected teenager with an internet connection to become a "script kiddie" hacker, leading to arms controls debates a decade later. The 2010s gave us blockchain technologies such as Bitcoin and other cryptocurrencies, which were built on defensive cyber technologies but whose lasting legacy remains the rise of ransomware attacks and online illicit marketplaces. The rise of AI presents a similar dilemma for regulators. A blanket export ban on advanced AI models is likely to be counterproductive. Open-source AI models such as DeepSeek lag only months behind the most advanced models of OpenAI and Anthropic. Recent research suggests that much of that gap can be closed by pairing less powerful AI models with complementary technologies. Defenders should therefore assume their adversaries already have access to AI on par with that used for cyber defence. Only by investing in strong foundations can they hope to escape the cat-and-mouse AI cyber arms race.

Five Eyes spooks warn AI means infosec incidents can become 'major operational and financial crises'

The leaders of intelligence agencies from the Five Eyes nations - Australia, Canada, New Zealand, the USA and the UK - have together issued strongly worded advice calling for leaders to nail cybersecurity basics or fall victim to ruinous AI-powered attacks. "The rapid pace of frontier AI development means cyber risk assumptions can become outdated in months, not years," the advice warns, and calls for organizations to take rapid action to ensure their defenses remain potent. "While AI will help us improve cyber defence over time, it also accelerates the speed, scale, and sophistication of cyber threats," the advice adds. "Frontier AI models are anticipated to exceed current industry expectations, fundamentally transforming both offensive and defensive cyber capabilities. The timeline is not years, it is months." After all that scary stuff, the spook bosses offer some antidote: "Cyber resilience is integral to advancing business continuity, market confidence, and long-term value." And how might one achieve that resilience? The Five Eyes have four suggestions: * Understand and assess risk, readiness and accountability * Prioritize foundational cyber security practices and controls * Empower cyber leaders with authority and resources * Stay actively engaged as threats and guidance evolve "Cyber risk can no longer be treated as a purely technical issue," the advice points out. "This is a core business risk and leadership responsibility," because breaches are inevitable and "Breaches will occur. Preparedness helps you contain them quickly and prevent escalation into major operational and financial crises." The intelligence chiefs therefore want organizations to test their cyber resilience rigs. "It is not enough to have controls," they write. "Leaders must be confident those controls will perform during a real incident. This requires reassessing long-standing trade-offs and using AI deliberately to strengthen defence - not just improve efficiency." That last sentence is a rare moment of optimism in the advice and precedes a section in which the intelligence bosses observe "Organizations that integrate AI tools into their security operations can detect vulnerabilities earlier, improve software quality, monitor unusual behaviour, and respond faster to incidents - reducing both the cost and impact of incidents." Readers of The Register might find this advice a little quaint given that infosec vendors have for years blathered on about the need for boards and bosses to take cyber seriously. It's also been a couple of years since it became apparent that generative and agentic AI can fuel new and unusually potent cyber-attacks. Interest in that idea spiked in the eleven weeks since Anthropic revealed the existence of its powerful flaw-finding Mythos model and hid it behind a regwall lest criminals use it to swiftly slice holes in important software. The Five Eyes bosses address their advice to "leaders" - presumably bosses of substantial organizations - who may not have watched the Mythos mess unfurl while they worried about a global energy crisis kicking holes in their supply chains. The good news is that the spy bosses don't think leaders need to learn a lot to cope with the advent of AI, as their advice suggests five practical actions they rate as "not new," but "now urgent to reduce not only technical risk, but also operational, financial and reputational exposure." For the record, those actions are: 1. Reduce your attack surface: Limit unnecessary system access and external connectivity. Challenge whether systems need to be exposed at all and isolate those that do not. 2. Accelerate patching processes: AI is shortening the time between vulnerability discovery and exploitation. Delays in patching increase risk, especially for operational systems with long update cycles. Prioritize security updates accordingly to manage risks. 3. Address legacy systems: Unsupported systems are easy targets. They are not just technical debt, they are strategic liabilities. 4. Review and strengthen identity and access controls: Limit who can access critical systems. Enforce strong authentication and regularly review permissions. 5. Prepare for incidents before they happen: Test response plans, train and prepare teams, and assume breaches will occur. Focus on fast containment and recovery. Take us, and this, to your leaders, dear readers. ®

Five Eyes alliance warns frontier AI cyber threats are 'months' away

A joint statement from five intelligence services says the next wave of AI will reshape offensive hacking, and that defenders are short of time. The Five Eyes intelligence alliance has issued a joint warning that the next generation of artificial intelligence is poised to supercharge offensive hacking, and that the window to prepare for it is closing fast. In a coordinated statement, the agencies of the United States, the United Kingdom, Canada, Australia, and New Zealand said urgent action was needed, and put a strikingly short clock on the threat. "Frontier AI models are anticipated to exceed current industry expectations, fundamentally transforming both offensive and defensive cyber capabilities," the statement read. "The timeline is not years, it is months." The agencies went on to warn that AI models capable of causing serious cyber harm are themselves only "months away" from being publicly available, a compression of the usual government risk horizon into something close to the present tense. Much of what the alliance flagged is the unglamorous machinery of how organisations get breached. The statement singled out legacy systems, slow patching cycles, unnecessary internet connectivity, weak identity and access controls, and a lack of pre-incident planning as the weaknesses that more capable AI will be quick to find and exploit. None of these are new problems; the argument is that AI will industrialise the exploitation of them, shrinking the time between a vulnerability becoming known and an attacker reaching it from weeks to something far shorter. A flaw that once took a skilled human team days to weaponise, the agencies suggest, could soon be turned into a working exploit by a model in a fraction of that time. That much of the underlying advice is familiar was, in a sense, the point. The bulk of the statement restated core cybersecurity hygiene, patch quickly, do not put systems online unless you need to, lock down who can reach what, the sort of guidance defenders have heard for years. The agencies also pressed defenders to turn the same technology back on the problem, urging organisations to use AI "to strengthen defence," for example by finding weaknesses sooner or responding to incidents faster. That framing mirrors a year in which the line between attacking and defending tool has grown thin: Google researchers used an AI system to surface a live zero-day exploit, and Anthropic has documented models that can uncover serious software vulnerabilities of the kind that keep banks awake. The warning lands amid a broader scramble to organise defences before the capability gap widens. Governments and vendors have been signing cross-border cyber partnerships, and the criminal use of AI is already visible at the edges, with researchers tracking AI-assisted crypto thefts attributed to North Korean operators. The Five Eyes statement effectively tells the rest of the field that the same tooling is about to become broadly available. The alliance was sounding an unusually loud siren while pointing organisations back towards basic discipline, an acknowledgement that most damage still flows through doors that were left unlocked. What the statement did not include was a fixed deadline or any regulatory mechanism, leaving the response to individual organisations and national agencies. Nor did it name particular AI labs or models, keeping the warning general rather than singling out any developer. For defenders, the practical takeaway is uncomfortable in its simplicity: the advice has not changed, but the time to act on it, by the alliance's own reckoning, is now measured in months rather than years.

Top Intel Agencies Say AI-Driven Cyber Catastrophes Are Imminent: 'The Timeline Is Not Years, It Is Months'

In a rare joint statement, the Five Eyes intelligence alliance -- the United States, Australia, the United Kingdom, Canada, and New Zealand -- warned on Monday that the cybersecurity threats posed by advanced AI models are approaching a critical point. "As the leaders of the Five Eyes cyber security agencies, we are united in our call to action: the evolving landscape of artificial intelligence (AI) is rapidly transforming cyber risk, and we must act swiftly to remain ahead," the alliance said in a joint statement signed by the intelligence chiefs of all five countries, including the United States' David Imbordino, who leads the National Security Agency's cybersecurity directorate, and Nick Andersen, who is the acting director of the Cybersecurity and Infrastructure Security Agency (CISA). "The timeline is not years, it is months." In the letter, the leaders say that developments in AI have been accelerating the "speed, scale, and sophistication of cyber threats" by lowering barriers for bad actors and shrinking the window between the discovery of a software vulnerability and its exploitation. "Cyber risk can no longer be treated as a purely technical issue. This is a core business risk and leadership responsibility," the letter reads. "Breaches will occur. Preparedness helps you contain them quickly and prevent escalation into major operational and financial crises." To help address the risks, the Five Eyes are urging leaders to limit unnecessary system access and external connectivity, avoid delays in patching vulnerabilities by prioritizing security updates, test response plans for potential breaches, strengthen identity authentication, and limit user access to critical systems. The group also urged organization leaders to integrate AI into their security operations. "Organisations that integrate AI tools into their security operations can detect vulnerabilities earlier, improve software quality, monitor unusual behaviour, and respond faster to incidents - reducing both the cost and impact of incidents," the group wrote. The statement comes at a tense time for cybersecurity. Earlier this year, Anthropic announced a new AI model called Mythos, which it said was so scary good at cracking software vulnerabilities that access could only be granted to select organizations and governments. Promptly after Anthropic's limited deployment of its spooky new, allegedly privacy-shattering model, OpenAI came forward with a model of its own with similar premises. According to reports from the few organizations that have gained access, the Mythos model is able to bypass Apple's notoriously tough-to-crack operating system and completely take over a corporate system in six out of 10 attempts. After announcing it had begun its long-awaited IPO process, Anthropic first expanded access to the Mythos model before releasing an allegedly safer defanged version to the public called Claude Fable 5. That model was not up for too long before the Trump administration intervened and forced Anthropic to suspend foreign nationals' access to both Fable 5 and Mythos, citing national security concerns. The ban included all foreign nationals living in or outside of the United States, including the company's own employees. To ensure compliance, Anthropic disabled access to both models for all users. The anticipated impact of these next-generation AI models is quickly becoming a major topic of discussion in global politics. Last week, AI company chiefs like Anthropic's Dario Amodei, OpenAI's Sam Altman, and Google DeepMind's Demis Hassabis were in attendance at the annual G7 Summit, taking seats at the same table as leaders of some of the most powerful governments in the world to discuss, among other things, the cyber risks of their models. While new AI models continue to put pressure on cybersecurity agencies around the world, the United States is facing another crisis of its own. Shortly after President Trump took office in January 2025, the nation's top cybersecurity agency, CISA, lost a third of its workforce to layoffs initiated by the administration. Even though Trump was semi-responsible for the agency's creation back in 2018 during his first term, he has since turned against it after officials refused to back his voter fraud claims in the 2020 presidential election that he lost to former President Joe Biden. In his second term, Trump has proposed more than $250 million in budget cuts to the organization. Last month, the agency was also involved in a rather embarrassing cybersecurity incident in which investigative journalist Brian Krebs found that CISA had left information like plaintext usernames and passwords for internal systems on GitHub, possibly for about six months. With AI models advancing rapidly and governments themselves admitting that they are proving to be a cybersecurity liability, it will be interesting to see how CISA, crippled under the Trump administration's attacks, will respond to these threats. So far, things are not looking particularly great, especially if you consider a recent report that the agency just gained full access to the model only two weeks ago.

Once, cyber-attacks required great skill. AI is changing that | Bruce Schneier

Modern AI systems are, in effect, a universal adviser to help people do harmful things. We'll need to harness AI for defense, too Earlier this week, national security agencies from the Five Eyes - that's the rich, English-language-speaking countries club - jointly released a statement warning of the increasing cyber risks of AI models: in particular, their ability to autonomously hack into systems and networks. The statement was more measured than some of the breathless headlines about it, and the advice they gave is pretty much the standard advice everyone gives - albeit with newfound urgency. Internet risks are nothing new, and cyber-attacks - both large and small - have been a significant issue since long before the current crop of generative AI models. What's been changing over the decades, and what AI is changing even faster, is the gap between skill and ability. For most of human history, the two terms were synonymous - but computers have decoupled them. As the gap between the two expands, humans empowered with these AI tools can do more: more writing, more research, more analysis and also more damage than ever before. These models can, with little detailed direction, autonomously hack into networks, steal data, deploy ransomware and destroy systems. And to the extent there is a solution, it's going to involve harnessing AI for the defense. In 1998, seven people from the hacker group L0pht testified before Congress. They told a mostly clueless Senate committee that they could take down the internet in 30 minutes. That was partly real and partly bravado, but it illustrates an important point: hacking into systems, stealing data and causing damage all required skill. Contrast the L0pht hackers with hackers derided as "script kiddies". They didn't understand computers, or security. Instead, they used hacker tools written by others. Their actions required minimal skill and even less knowledge. But once those hacking tools became widespread, the number of potential attackers increased. That number has continued to increase, as quality and availability of prewritten attack tools has grown. And it is growing dramatically with AI. Today's AI systems - not just the frontier models, but most of them - are capable of carrying out cyber-attacks automatically. They all do better in the hands of skilled attackers, but increasingly they are able to act autonomously with only minimal prompting. The thing about people with ability but no skill is that they are often outsiders, not part of any professional community, and not bound by any rules or norms. This phenomenon is much more general than in cybersecurity. Any doctor can tell you how to untraceably poison someone, and many virus researchers know how to create a bioweapon. Any bridge engineer can tell you how to place explosives to blow a bridge up. The reason that murderous doctors and terrorist engineers are so rare is that the lengthy process of acquiring those skills also instills a moral and ethical code. If every random person has access to good poisoning advice, that puts us all in danger. Modern AI systems are, in effect, a universal adviser to help people do harmful things. And while the current AI megacorporations are trying to build guardrails to prevent people from asking questions whose answers will enable the questioner to do harm, that's not going to work in the long term. Smaller, cheaper, open-source models, including models that can run on people's computers, and especially groups of models that run in concert with each other, are just as good as the frontier models from companies like OpenAI and Anthropic. And they continue to get better. These models will be passed around from person to person, like script kiddie hacker tools, and they won't have any such guardrails. Instructing AI models to spy on people and report any malicious prompts to the authorities fails for similar reasons. The megacorporations can do that, but the locally run open source models won't. This could buy us a few months at best. A third possibility is to somehow make the models themselves unable to hack into computers, create bioweapons or do anything else that might harm people or society. That won't work, for the same reason we can't teach doctors how to treat poisonings without also teaching them how to poison. It's the same knowledge. It's the same with construction and demolition. And it's the same with cybersecurity. We want these AI models to be able to review computer code, find vulnerabilities and automatically fix them. The benefit to our collective security will be enormous. Unfortunately, the same knowledge can be used for attacks. Where this leaves us is in a world of increased volatility. Super-powered humans with AI assistants will be able to do both wonderful and horrible things. This brings us back to the Five Eyes statement. Everything they recommend is something security professionals have been recommending for years, if not decades. They are things talked about at that congressional hearing back in 1998, titled "Weak computer security in government: Is the public at risk?" Even the Five Eyes admitted last week that their security advice is not new, only more urgent. What's new is how fast things are changing: "The rapid pace of frontier AI development means cyber risk assumptions can become outdated in months, not years. We must act before and be prepared to adapt and withstand evolving threats." The Five Eyes point to AI technology - not necessarily chatbots, but AI more generally - being used to strengthen every aspect of defense, to "detect vulnerabilities earlier, improve software quality, monitor unusual behavior, and respond faster to incidents - reducing both the cost and impact of incidents". Excellent advice from the Five Eyes security agencies. We need to do this with every risk that AI heightens, not just cybersecurity.

Behind the Curtain: America's AI lead, security in peril

Why it matters: Yes, Anthropic's Mythos model is the most cyber-lethal threat in the world. But OpenAI is close here in America. And China and Japan, using much cheaper models, have gotten closer, faster than intelligence agencies anticipated. * Five Eyes, composed of the U.S., U.K., Canada, Australia and New Zealand, is considered the world's most comprehensive and powerful spy network. The big picture: President Trump told Marc Caputo on "The Axios Show" that "we're beating China by a lot" on AI -- but that lead, which U.S. leaders and businesses have been banking on, is eroding. Three new disruptions show just how fast it's happening: Between the lines: China's open-source models are gaining ground fast. Z.ai's GLM-5.2 is the buzziest right now. * An LLM leaderboard by Artificial Analysis, a benchmarking company, puts GLM-5.2 alongside OpenAI's GPT-5.5, at about a fifth of the cost to run. When it comes to coding, Arena's web development ranking has the Chinese model second only to Fable -- making it the best-performing model you can actually use right now. Alex Stamos, former Facebook chief security officer, told Axios Future of Cybersecurity that it's quite possible the Chinese "have things privately that are really, really good. [It] is arrogant and foolish of us to think that just because we're American that we've got the best stuff." * He added that Chinese military hackers are likely "laughing hilariously right now at the Americans fighting between themselves and cutting each other off left and right." Zoom out: It's Europe, too. Domyn -- an AI company based in Milan, Italy -- announced last week that its Europa project is a frontier open-source AI model that will support all 24 official languages of the European Union. * Domyn (formerly iGenius, which was described as the Ferrari of AI) collaborated with Nvidia to build Colosseum, billed as Europe's largest AI supercomputer. What we're watching: The Five Eyes call to action said a "whole-of-society response is required" to address accelerating cyber risk. * "Boards and executives should ensure cyber resilience is in place and works under pressure," the bulletin says. "It is not enough to have controls. Leaders must be confident those controls will perform during a real incident. This requires reassessing long-standing trade-offs and using AI deliberately to strengthen defense -- not just improve efficiency." The bottom line: America's AI lead is real but shrinking. Every move to protect it only hands rivals another reason to route around it -- all while the capabilities that have Five Eyes on edge are already loose, downloadable and impossible to recall. * Axios' Shane Savitsky and Sam Sabin contributed reporting. 📈 If you're a CEO or on a CEO's team: Ask to join Jim's new weekly Axios C-Suite newsletter.

'Act now': Five Eyes warns that AI models specialized for cyber attacks are only months away

A whole-of-organisation and whole-of-society response is required, Five Eyes is saying * Five Eyes alliance warned frontier GenAI models will enable advanced cyberattacks against businesses and governments within months * Statement stressed cyber risk is now a leadership and business continuity issue, requiring whole‑of‑society response * Comes amid concerns over Anthropic's Mythos Preview and other models already showing offensive potential despite guardrails In just a few months, high-end Generative Artificial Intelligence models (GenAI) will be capable of running cyberattacks on big businesses and government organizations, Five Eyes is warning. The Five Eyes is an intelligence-sharing alliance between the United States, United Kingdom, Canada, Australia, and New Zealand. Formed after the Second World War, it allows the five countries to closely cooperate on intelligence and matters of national security. Earlier this week, Five Eyes issued a new warning, saying that AI will help improve cyber defense over time, but will also accelerate the speed, scale, and sophistication, of threats: "Frontier AI models are anticipated to exceed current industry expectations, fundamentally transforming both offensive and defensive cyber capabilities. The timeline is not years, it is months," the warning reads. "In this environment, cyber resilience is integral to advancing business continuity, market confidence, and long-term value." All hands on deck Five Eyes is now saying that the industry needs all hands on deck to address what's increasingly becoming a burning issue: "A whole-of-organisation and whole-of-society response is required," it said. "Cyber risk can no longer be treated as a purely technical issue. This is a core business risk and leadership responsibility." In early April, news broke that Anthropic's latest AI model, Mythos Preview, was so good at exploiting software vulnerabilities, that the company could not release it to the public. Instead, it only shared it with a handful of US enterprises, to give them a head start against threat actors. While skeptics said it was nothing more than a publicity stunt, similar to what OpenAI pulled off with ChatGPT 2.0, companies that used it (for example, Mozilla), confirmed that it was, indeed, powerful enough that it needs to be kept in check. Even models available today, despite all the guardrails, are being regularly leveraged by bad actors in different cyberattack scenarios. Via The Guardian Follow TechRadar on Google News and add us as a preferred source to get our expert news, reviews, and opinion in your feeds.

AI on pace to bypass cybersecurity systems in months, not years, "Five Eyes" spy partners warn

The most advanced artificial intelligence models are improving quickly enough to outsmart prevailing cybersecurity know-how within months, the Five Eyes spy agency alliance has warned. The risk posed by AI-enhanced hacking is in the spotlight in the wake of startup Anthropic saying in April that its cutting-edge Mythos models had unprecedented abilities to find software vulnerabilities. The security agencies of Britain, the United States, Australia, Canada and New Zealand urged governments and businesses to act swiftly to prepare themselves as AI evolves. "The rapid pace of frontier AI development means cyber risk assumptions can become outdated in months, not years," said a joint statement dated Monday. AI "lowers barriers for malicious actors and increases the speed and complexity of attacks," the Five Eyes advisory said. "Breaches will occur. Preparedness helps you contain them quickly and prevent escalation into major operational and financial crises." To improve cyber defenses, organizations should integrate AI tools into their security operations, update old systems and limit access to critical systems among other steps, they said. Anthropic this month suspended access to Mythos 5 and a restricted version called Fable 5 to comply with a U.S. national security order. Just days after publicly launching Fable 5, the company said it had received a government directive banning all foreign nationals from accessing the two models. The intervention is striking for a White House that has otherwise pushed to loosen AI oversight -- even moving to block states from writing their own rules.

AI models that can take down governments and business months away, rare Five Eyes statement warns

Signal agencies in Australia, the US, the UK, New Zealand and Canada sound alarm after Trump blocks foreign nationals from Anthropic's Fable AI model Powerful AI models capable of taking down governments and businesses are mere months away, cyber intelligence agencies for the Five Eyes have warned in a rare joint statement, urging leaders to "act now". The surprising public intervention by signals agencies for Australia, the US, the UK, New Zealand and Canada comes after the Trump administration earlier this month decided to block "foreign nationals" from using a much-hyped AI model built by tech company Anthropic, called Fable. The statement, issued late Monday night Sydney time, said while AI "would help us improve cyber defence over time, it also accelerates the speed, scale, and sophistication of cyber threats". "Frontier AI models are anticipated to exceed current industry expectations, fundamentally transforming both offensive and defensive cyber capabilities. The timeline is not years, it is months," the warning by Five Eyes' agencies said. "In this environment, cyber resilience is integral to advancing business continuity, market confidence, and long-term value." The cybersecurity agencies said the leaps in AI models showed the technology would lower barriers for bad actors and increase the speed and complexity of attacks. "A whole-of-organisation and whole-of-society response is required," the statement continued. "Cyber risk can no longer be treated as a purely technical issue. This is a core business risk and leadership responsibility." While no AI models or companies are by mentioned in the statement by name, many around the world have their eyes on Anthropic's advanced tier of tools. One of the major tech company's latest inventions is called Fable 5, a supposedly more community-friendly version of Mythos - a powerful AI model released earlier this year capable of detecting vulnerabilities in cyber systems that is only available to vetted organisations and companies because of concerns it could be exploited for bad. Both of Anthropic's models were suspended for use by "foreign nationals" in June by the US government, which cited advice by national security authorities. Olivia Shen, an expert in national security and AI at the University of Sydney's United States Studies Centre, said much of the world was focused on what happens next for Anthropic but there could be many more powerful AI models not far off on the horizon. "I think we have to anticipate that the next Mythos or the next Fable is just around the corner," Shen said. "We can only see what's been released, but there could be other models being developed by the likes of China, or other states and other actors and companies, that are just as advanced." In March, the Albanese government signed Anthropic as the first company on to its national AI plan. The non-binding memorandum of understanding means companies agree to share details of AI progress with the government and "promote safety". The government's national plan promotes a light-touch approach on regulating the sector in a bid to capture economic and productivity benefits from the technology.

China's AI advances collide with U.S. safety debate

Why it matters: The world is only months away from AI models dramatically accelerating cyber threats, Five Eyes leaders warned Monday. * Yet preparations are being slowed by Washington infighting and industry-wide confusion over how to measure AI risk. Driving the news: A new Chinese open-source model, GLM-5.2, rocked the internet this weekend with its ability to match the agentic capabilities of models like Anthropic's Opus 4.8 -- garnering praise from Silicon Valley elites and raising questions about just how quickly China will close the gap. * At the same time, the Trump administration is still debating the best way to release Anthropic's Fable 5 and Mythos 5 models over safety and national security concerns. The big picture: The biggest disagreement in AI security right now isn't whether China is catching up -- it's how quickly. * Stanford's AI Index Report suggests Chinese models rapidly caught up in quality over the past year and have largely erased the U.S. advantage. * Former White House AI czar David Sacks said just this month that the U.S. only has a six- to nine-month lead on China. * But others argue that benchmark gains alone don't mean China has solved the compute and infrastructure challenges needed to truly compete at the frontier. What they're saying: "It is quite possible they have things privately that are really, really good, and [it] is arrogant and foolish of us to think that just because we're American that we've got the best stuff," Alex Stamos, former Facebook security leader, told Axios last week. * He added that Chinese military hackers are likely "laughing hilariously right now at the Americans fighting between themselves and cutting each other off left and right." Threat level: Officials and lawmakers fear China could use powerful AI systems to beef up surveillance, cyber operations and military decision-making. * China's embrace of open-source models could also make its AI ecosystem more economically attractive globally, particularly for companies seeking alternatives to expensive U.S. frontier models. Between the lines: Much of the debate isn't about what Chinese models can do today. It's about the risk of being surprised tomorrow. * One open-source security researcher, who requested anonymity because he was not authorized by his employer to speak publicly, told Axios he uses the frontier models to understand how capabilities like persuasion, social engineering and vulnerability discovery are evolving. * He's worried that as Chinese models improve, restricting access to cutting-edge U.S. systems could leave defenders with fewer ways to anticipate what's coming next. Yes, but: Not everyone believes China's progress represents an imminent threat to U.S. AI leadership. * China lacks the "bleeding-edge chips" and vast amounts of data needed to develop a competitive frontier AI model, Pukar Hamal, founder and CEO of SecurityPal AI, told Axios. * "Who has access to the most chips and most data? It's American companies, so far," he added. Reality check: Researchers can already find many of the bugs that advanced models like Mythos are finding without using the models, which often are more costly and difficult to gain access to. * AI-powered security firm Aisle claimed last week that its agentic capabilities are outperforming Mythos in several tests already. What to watch: OpenAI just made its cyber model, GPT-5.5-Cyber, more permissive and capable. On key benchmarks, it outperforms Anthropic's Mythos 5.

Intelligence agencies warn AI models could launch crippling cyberattacks in months

A group of intelligence agencies from across the world, including the U.S., warned on Monday that artificial intelligence is "rapidly transforming" the cybersecurity risks, urging global leaders to "act swiftly" to stay ahead of malicious actors. The Five Eyes group, including the U.S., U.K., Canada, Australia and New Zealand, said in a joint statement that AI's impacts on defensive and offensive cyber threats is not happening in years, but months. "While Al will help us improve cyber defence over time, it also accelerates the speed, scale, and sophistication of cyber threats," the countries wrote, adding "AI is not a future consideration -- it is already here." AI firms have long warned of how the technology could empower bad actors to hack faster, cheaper and at a broader scale, but concerns ramped up this year amid these firms' new cybersecurity models like Anthropic's Mythos model. Anthropic held back the full release of the Mythos model earlier this year, saying it was too dangerous for public use. The firm maintains Mythos is able to spot decades-old vulnerabilities, making it easier for governments, software and infrastructure to patch things up, but also easier for bad actors. "[AI] lowers barriers for malicious actors and increases the speed and complexity of attacks, shrinking the window between vulnerability discovery and exploitation ever more quickly" the Five Eyes grouping said. "At the same time, AI offers powerful tools to strengthen defence." The warming is directed at small and medium businesses, large organizations and infrastructure, and governments, all of which they suggest are at risk to these threats. They urged leaders to accelerate their patching processes, address unsupported legacy systems and "prepare for incidents before they happen. Leaders should implement secure-by-design and secure-by-default as standard practice, meaning security is implemented at the creation of any software or system. Washington has spent the last few months debating how the White House should handle the cybersecurity risks of newer AI models, as the administration tries to balance AI safety concerns with its longtime commitment to light-touch regulation. Trump signed an order earlier this month laying out a voluntary testing process in which AI labs can provide the government with their models up to 30 days ahead of release to test for certain risks. While the administration emphasized testing was not mandatory, some predicted at the time these assurances would not be enough. Just weeks later, the administration issued a directive prompting Anthropic to pull newest Fable and Mythos models, sparking intense backlash from the AI policy community.

The Alarming Reason 5 Intelligence Agencies Just Issued a Next-Gen AI Warning

AI is evolving so rapidly, it could pose a cybersecurity threat to governments and businesses within months. That's according to the Five Eyes intelligence partnership, which released a joint call-to-action on Monday that urges business leaders to secure their operations now in the interest of "business continuity, market confidence, and long-term value." "Frontier Al models are anticipated to exceed current industry expectations, fundamentally transforming both offensive and defensive cyber capabilities. The timeline is not years, it is months," the bulletin reads. The Five Eyes includes the FBI, the UK's MI5, and intelligence agencies from Australia, Canada, and New Zealand. It wasn't clear what specifically prompted the communique, but it follows the cautious rollout of Anthropic's Claude Mythos model in preview mode to a limited number of trusted organizations. Mythos is reportedly so powerful that it can "surpass all but the most skilled humans at finding and exploiting software vulnerabilities."

Frontier AI May Make Cyberattacks Faster, More Dangerous, Five Eyes Says

Cyber security an operational issue, faster software patching is critical A joint warning was issued by intelligence and cybersecurity agencies from Australia, Canada, New Zealand, the UK, and the US on Monday. In a statement, the alliance, commonly known as the Five Eyes, said that AI has the potential to dramatically accelerate cyberattacks in the coming months. The agencies have warned against frontier AI models that have developed the capability of both offensive and defensive actions sooner than previously anticipated, claiming that cybersecurity cannot be treated as a purely technical issue anymore. AI to Transform Cyber Threats Within Months In a three-page joint statement, the Five Eyes cybersecurity agencies said that frontier AI models are advancing rapidly. It is expected to transform the cybersecurity landscape within months rather than years. "Frontier AI models are anticipated to exceed current industry expectations, fundamentally transforming both offensive and defensive cyber capabilities. The timeline is not years, it is months," the agencies said. A frontier AI model, notably, is a cutting-edge general-purpose AI model that can deliver the highest level of performance on fronts like coding, general knowledge, multimodality, and reasoning at any given time. The "frontier" status is said to be temporary as AI research moves incredibly fast. Thus, the boundary line continues to shift as new systems are released by AI firms. The warning against frontier models potentially accelerating cyberattacks was signed by cybersecurity authorities from the five member nations. These include the US Cybersecurity and Infrastructure Security Agency (CISA). While it did not specifically mention an AI model, a concern was raised towards the increasingly capable systems that could help malicious actors automate attacks, discover vulnerabilities faster, and scale cyber operations more efficiently than previously possible. As per the agencies, AI can already reduce the time between discovering a software flaw and exploiting it, making traditional patching timelines inadequate. This is said to especially affect organisations that leverage legacy infrastructure or critical systems with long update cycles. Organisations have been advised to reduce their attack surface by accelerating software updates, moving away from unsupported systems, regularly testing incident response plans, and strengthening identity and access management controls. Five Eyes also noted that security breaches should be considered as inevitable events and should be treated as such. "Cyber resilience is not an IT issue -- it is central to operational continuity and market trust," the Five Eyes agencies said. They added that leaders who act now will be better positioned to manage evolving cyber risks, while those who delay may face increasing operational, financial, and reputational exposure. On the defensive front, cybersecurity agencies emphasised that AI can be a powerful tool if responsibly deployed, as it can identify vulnerabilities earlier, improve software quality, monitor suspicious activity, and respond to incidents more quickly.

'Five Eyes' intelligence alliance warns that new AI models pose urgent cyber risk

The intelligence alliance commonly known ⁠as the "Five ⁠Eyes" said in a three-page statement that, "Frontier AI models are anticipated to exceed current industry expectations, fundamentally transforming both offensive and defensive cyber capabilities. The timeline is not years, it is months." Cutting-edge artificial intelligence technology is poised to supercharge offensive hacking capabilities and urgent action is needed to face up to the threat, U.S., British, Canadian, Australian and New Zealand officials said on Monday. The intelligence alliance commonly known ⁠as the "Five ⁠Eyes" said in a three-page statement that, "Frontier AI models are anticipated to exceed current industry expectations, fundamentally transforming both offensive and defensive cyber capabilities. The timeline is not years, it is months." The statement was light on details and mostly restated core cybersecurity advice, ⁠such as swiftly patching faulty software and not putting systems online unless necessary. The officials also urged ⁠defenders to use AI "to strengthen defence," for example by identifying weaknesses sooner or responding more quickly to incidents. The warning was another indication of officials' increasing concerns over models such as Anthropic's "Mythos" or OpenAI's "GPT-5.5-Cyber," which are said to allow users to quickly execute complex - and potentially devastating - hacks. Earlier this month, Anthropic was forced to disable a version of Mythos after the U.S. government ordered it to suspend ⁠access to the models for foreign nationals over alleged national security concerns. Around the same time, the U.S. cyber defense agency CISA - which was among those cosigning Monday's statement - reduced the deadlines imposed on government officials to deal with serious digital vulnerabilities in their networks to three days, citing AI threats.

Five Eyes Alliance Warns AI Could Supercharge Cyberattacks

The Five Eyes intelligence alliance has issued a stark warning that advances in artificial intelligence (AI) could dramatically accelerate cyberattacks in the near future, arguing that organizations have only months -- not years -- to prepare for a rapidly changing threat landscape. Five Eyes Says Frontier AI Will Transform Cyber Warfare In a joint statement released Monday, intelligence and cybersecurity officials from the U.S., U.K., Canada, Australia and New Zealand said next-generation AI systems are expected to "fundamentally" reshape both offensive and defensive cyber capabilities. "Frontier AI models are anticipated to exceed current industry expectations," the agencies said, warning that "the timeline is not years, it is months." The statement reflects growing concern among Western governments that increasingly capable AI models could help attackers identify vulnerabilities, write malicious code and automate complex cyber operations at unprecedented speed. AI Models Draw Growing National Security Scrutiny The warning comes amid heightened debate over the cybersecurity implications of advanced AI systems developed by companies such as Anthropic and OpenAI. According to the report, concerns have grown that cutting-edge models could lower the barriers to conducting sophisticated cyberattacks. Earlier this month, Anthropic reportedly suspended access to a version of its Mythos model after a U.S. government directive tied to national security concerns involving foreign users. At the same time, the U.S. Cybersecurity and Infrastructure Security Agency, or CISA, reportedly shortened deadlines for federal agencies to remediate critical vulnerabilities to just three days, citing the growing pace of AI-enabled threats, Reuters reported. Disclaimer: This content was partially produced with the help of AI tools and was reviewed and published by Benzinga editors. Photo courtesy: Golden Dayz / Shutterstock Market News and Data brought to you by Benzinga APIs To add Benzinga News as your preferred source on Google, click here.

'Five Eyes' intelligence alliance warns that new AI models pose urgent cyber risk

WASHINGTON, June 22 (Reuters) - Cutting-edge artificial intelligence technology is poised to supercharge offensive hacking capabilities and urgent action is needed to face up to the threat, U.S., British, Canadian, Australian and New Zealand officials said on Monday. The intelligence alliance commonly known as the "Five Eyes" said in a three-page statement that, "Frontier AI models are anticipated to exceed current industry expectations, fundamentally transforming both offensive and defensive cyber capabilities. The timeline is not years, it is months." The statement was light on details and mostly restated core cybersecurity advice, such as swiftly patching faulty software and not putting systems online unless necessary. The officials also urged defenders to use AI "to strengthen defence," for example by identifying weaknesses sooner or responding more quickly to incidents. The warning was another indication of officials' increasing concerns over models such as Anthropic's "Mythos" or OpenAI's "GPT-5.5-Cyber," which are said to allow users to quickly execute complex -- and potentially devastating -- hacks. Earlier this month, Anthropic was forced to disable a version of Mythos after the U.S. government ordered it to suspend access to the models for foreign nationals over alleged national security ?concerns. Around the same time, the U.S. cyber defense agency CISA -- which was among those cosigning Monday's statement -- reduced the deadlines imposed on government officials to deal with serious digital vulnerabilities in their networks to three days, citing AI threats. (Reporting by Raphael Satter in Washington; Editing by Matthew Lewis)