Nvidia forms Open Secure AI Alliance with Microsoft, but OpenAI, Google and Anthropic sit out

Nvidia, Microsoft launch open AI security alliance - without OpenAI, Google, or Anthropic

Nvidia on Monday said it is joining forces with Microsoft, SpaceX, IBM, and other tech companies to build and share open-source AI security tools. The new Open Secure AI Alliance said open tools are required to effectively defend against attacks from frontier models. The initiative is a direct response to mounting concerns over the safety of advanced AI systems after a rogue OpenAI model escaped containment and attacked another company during testing. That company, Hugging Face, said it was forced to use a Chinese open-weight model to defend itself due to the strict safety guardrails limiting the usefulness of top US models. Founding members include Palantir, OpenClaw, the Linux Foundation, Cloudflare, Cloudera, Dell, Cisco, Adobe, Siemens, and DoorDash. Conspicuously absent are leading US AI companies, including OpenAI, Google, and Anthropic. The alliance arrives amid growing tensions over whether the world's most capable AI models should remain open. Chinese companies have released increasingly powerful open-weight models, notably Moonshot AI's Kimi K3, challenging the strategy pursued by US labs that have largely kept frontier systems closed and proprietary. Nvidia and its partners argue that securing AI requires access to both closed and open models, stressing that defenders need the tools to counter emerging threats. The announcement also follows reports that the Trump administration considered restricting access to cutting-edge Chinese models and an industry movement -- again spearheaded by Nvidia -- defending the need for openness in AI. Google and OpenAI signed that letter belatedly, too, though Anthropic remains absent.

Is open source the answer to rogue AI agents? Nvidia's new alliance says yes

Follow ZDNET: Add us as a preferred source on Google. ZDNET's key takeaways * Nvidia's new open-source alliance aims to democratize security solutions. * AI is often the best tool for combating AI-driven attacks. * Agent harnesses are a key component of better security. AI agents are behind a steady string of security incidents this year. Nvidia thinks a new partnership built on open-source software is the answer. On Monday, the company announced the Open Secure AI Alliance, which it said "will work to remediate and disclose vulnerabilities using open technologies." The announcement is something of a response to Project Glasswing, the security alliance Anthropic spearheaded around its highly capable Mythos 5 model. Nvidia's initiative aims to further democratize AI security tools by focusing on open-source software rather than reserving access to a proprietary model for a few major companies. Also: OpenAI's attack agent did exactly what it was told - just more relentlessly than expected Nvidia said the Alliance was spurred by last week's Hugging Face incident, in which an OpenAI agent escaped a testing environment and infiltrated Hugging Face, stealing credentials and demonstrating what OpenAI said was an "unprecedented" outcome. Nvidia argued in its announcement that because AI tools themselves have become an effective way to remedy AI attacks, open-source security tools must be a reliable public good. "When closed AI tools -- unable to distinguish attackers from defenders -- blocked essential forensic analysis, Hugging Face ran the open-weight GLM 5.2 model on its own infrastructure to analyze more than 17,000 actions and contain the intrusion," Nvidia noted. Cloudflare, CrowdStrike, Adobe, IBM, the Linux Foundation, Microsoft, and ex-OpenAI executive Mira Murati's startup Thinking Machines Lab are among the first participants in the Alliance. Nvidia said its contribution to the effort will include new research on agent harnesses -- the infrastructure that turns an LLM into an agent by helping it act autonomously -- as well as open models, weights, and data. The security case for open source In the announcement, Nvidia specifically argued for open models (alongside closed models) as a solution to security incidents, countering the dominant narrative that open-source AI can be more easily hijacked. Also: How AI has suddenly become much more useful to open-source developers "Some argue that open models are inherently less safe because they can be misused for cyberattacks or modified to remove guardrails," the company said. "Those risks are real, but they do not disappear in closed systems, and simply keeping weights closed does not prevent determined attackers from seeking or exploiting powerful AI." When it comes to AI, open-source means open-weight; that is, the model's final parameters and biases are public (rather than closed, as with Anthropic's or OpenAI's). Having that information, which shapes a model's outputs, lets developers fine-tune models for their own needs. Also: 'Like handing out the blueprint to a bank vault': Why AI led one company to abandon open source A model is truly open-source, however, when its code and training dataset are publicly accessible, meaning anyone could access its building blocks and understand more thoroughly how it works. Given how lucrative powerful proprietary models can be for companies like Anthropic and OpenAI, there are few incentives for fully open-sourcing a model. A call for pro-open-source policy Nvidia also pointed to the overall characterization of open-source AI as a possible hindrance for cybersecurity efforts going forward. Also: Moonshot's open-source Kimi K3 model beats Anthropic's Fable 5 on this benchmark "It will be crucial to recognize open models, harnesses, and security tooling as defensive assets, not liabilities, in AI and cybersecurity policy," the company said in the announcement. "Blanket restrictions on open frontier AI systems would weaken defensive capacity and risk concentrating power, dependence, and vulnerability in a few closed providers." The Trump administration has been especially critical of open models from Chinese startups like Moonshoot and DeepSeek, which are often competitive with those from US labs, citing security risks. While those concerns are valid, they are also colored by worries that China will outpace the US in building the most sophisticated AI systems. Nvidia's call to regulators also touches on the administration's increasing involvement in Anthropic and OpenAI's model release timelines reagrding security issues. The government was central to the Fable 5 and Mythos 5 recall and gave OpenAI prior approval to release its latest model, GPT-5.6.

OpenAI, Google, and Anthropic absent from Nvidia-led Open Secure AI Alliance -- 30+ companies join security alliance after OpenAI agent breach

OpenAI, Google, and Anthropic, companies behind proprietary, "closed" AI models are conspicuously absent from the alliance. A coalition of over 30 tech industry leaders, including Nvidia, Microsoft, SpaceX, The Linux Foundation, Adobe, and Siemens has formed the "Open Secure AI Alliance" with the aim of building and distributing open source tools for AI safety and security, according to an official Nvidia blog post on Monday. The Nvidia-led coalition -- comprising a mix of infrastructure, cloud computing, cybersecurity, and enterprise software leaders -- will serve as a collaborative effort to develop open tools for identifying and patching AI vulnerabilities, sharing security frameworks, and establishing identity verification and audit standards across the AI software stack. Curiously, some of the biggest names in AI, including OpenAI, Anthropic, and Google, are absent from the list of members. "The world needs both closed and open models. For cybersecurity, open models and open harnesses are essential because they democratize defensive capabilities, increase transparency for defenders, enable cyber defense while protecting data, and complement frontier closed models with customizable, localized controls. Open source enables massively distributed community-driven and self-controlled defense - with no single point of failure," the announcement reads. Contributors across the alliance are currently building or offering various tools to create an open defense stack. The initiative was directly galvanized by the OpenAI HuggingFace security incident earlier this month in which an autonomous OpenAI test agent slipped out of its sandbox and breached the AI startup Hugging Face. During the incident, safety guardrails on several frontier closed models prevented developers from performing critical forensic analysis. Hugging Face eventually had to use GLM-5.2 -- an open-weight model from Beijing-based Z.ai -- running the model on its own infrastructure to analyze more than 17,000 actions and contain the intrusion. Based on the incident, the alliance contends that being unable to inspect, modify, or run a model locally -- impossible in closed systems but doable with open systems -- presents a fundamental weakness in relying exclusively on closed AI systems for cyber defense. The Open Secure AI Alliance therefore aims to give entities access to advanced open models, agent harnesses, and security tools that they can independently deploy and adapt, reducing dependence on any single provider while strengthening defenses across a multi-vendor AI ecosystem. "That is the mission of the Open Secure AI Alliance: to ensure defenders everywhere have open, frontier tools they can trust and control," the post says. Chinese models such as DeepSeek and the newly released Kimi K3 are open-weight, and are seeing growing adoption, including by U.S. companies, due to their open features. Meanwhile the Trump administration is reportedly gearing up to ban Chinese AI models over security concerns. The alliance acknowledges the potential risks of open source tools but argues that closed systems are not an outright solution. "Those risks are real, but they do not disappear in closed systems, and simply keeping weights closed does not prevent determined attackers from seeking or exploiting powerful AI," the announcement reads. Unlike regulators who have voiced concerns over open-source technology, the alliance urges policymakers to treat open-weight models as defensive assets rather than liabilities. It also argues that placing AI development solely in the hands of a few closed providers creates dangerous single points of failure. "The right response is not to deny defenders access to capable open systems. It is to pair openness with strong safeguards, clear rules against malicious misuse, rigorous evaluation and rapid remediation. Defenders need both frontier closed models and frontier open models, working together, so they can choose the right system for the job and ensure that transparency, adaptation and sovereign control are available wherever security demands them," the alliance contends. According to the announcement, "The Open Secure AI Alliance -- building on the leadership of the Linux Foundation's Akrites initiative and OpenSSF community work -- will work to remediate and disclose vulnerabilities using open technologies". Founding members include NVIDIA, Dell Technologies, Synopsys, Microsoft, IBM, Red Hat, CrowdStrike, Palo Alto Networks, Cloudflare, Hugging Face, Databricks, SpaceXAI, and The Linux Foundation. Conspicuously absent from the alliance are OpenAI, Google, and Anthropic, companies behind proprietary, "closed" AI models. Follow Tom's Hardware on Google News, or add us as a preferred source, to get our latest news, analysis, & reviews in your feeds.

NVIDIA & Others Form The Open Secure AI Alliance

"The recent Hugging Face security incident delivered a clear reminder: cyber defenders need open, frontier agentic systems for self-defense. When closed AI tools -- unable to distinguish attackers from defenders -- blocked essential forensic analysis, Hugging Face ran the open-weight GLM 5.2 model on its own infrastructure to analyze more than 17,000 actions and contain the intrusion. That incident showed a practical truth: when defenders cannot inspect, adapt and run advanced AI on their own infrastructure, their ability to respond is constrained at exactly the moment speed matters most. Companies and countries need open frontier defensive tools and techniques so critical industries can build security systems across a multi-vendor ecosystem and avoid single points of failure. That is the mission of the Open Secure AI Alliance: to ensure defenders everywhere have open, frontier tools they can trust and control. ... Defenders need both frontier closed models and frontier open models, working together, so they can choose the right system for the job and ensure that transparency, adaptation and sovereign control are available wherever security demands them."

Tech giants link hands to praise open AI models after OpenAI - Hugging Face attack

In the wake of OpenAI agents attacking Hugging Face, Nvidia has recruited a new posse of partners to promote open source models as the security solution the industry needs. The AI arms dealer announced the foundation, the Open Secure AI Alliance, in a blog post today, describing the mission of the group being "to ensure defenders everywhere have open, frontier tools they can trust and control." Partners in the group are numerous, ranging from established tech giants like Microsoft, Red Hat, HPE, IBM, and Adobe to newer groups like Palantir, SpacexAI, Hugging Face, and The Linux Foundation. What all the founding members have in common, Nvidia said, is that they agree open source AI models are a fundamental part of modern cybersecurity, just like prior open source tech has been for the infosec space. "The United States and its partners now face a choice in AI security: whether the defenses that protect our infrastructure will sit inside a few opaque systems or be built on open models, harnesses and tools that any defender can study, adapt and deploy," Nvidia said in the announcement. The claims in many ways echo the pleadings from tech industry heavyweights made in an open letter to US government regulators last week. That letter, signed by many of the same companies that are part of the founding OSAA cadre, essentially argues that regulators should ensure Anthropic, Google, and OpenAI don't end up with total control of the US AI market, and that open-weight models should be given a seat at the table, too. The new alliance is arguing that, not only do open-weight models need to be allowed to proliferate in the US, but they also need to be considered a fundamental part of the security puzzle. For those unfamiliar with the Hugging Face incident, a group of autonomous OpenAI agents, operating in a sandbox and stripped of guardrails to test their full capability to solve cybersecurity puzzles, exploited a pair of zero-days to escape and gain access to the internet. For some reason, the bots thought the solution to the problems they were posed could be found in Hugging Face systems, so they broke in and accessed a bunch of private information and hijacked some credentials. When Hugging Face turned to closed-source US frontier AI lab bots to examine the incident and help figure out what happened, those tools declined to help because they thought the data Hugging Face was trying to examine was itself malicious. Hugging Face turned to Chinese-made GLM 5.2, hosted on its own infrastructure, to figure things out. "That incident showed a practical truth," said Nvidia. "When defenders cannot inspect, adapt and run advanced AI on their own infrastructure, their ability to respond is constrained at exactly the moment speed matters most." Only open-source AI models, which China leads development on, can fill that role, the OSAA argues, and it's prepared to counter those who say open models are a threat: Just look at what happened last week and it's readily apparent that closed source models are dangerous too. The Alliance is pooling its efforts to give security pros access to essential open tools. Nvidia said that it's participating by releasing its Object-Oriented Agent project on GitHub, HPE is contributing its SPIFFE/SPIRE zero-trust AI identity framework, Hugging Face has handed its Safetensors transparent AI model weight formatting to the PyTorch Foundation, and SpaceXAI has open-sourced Grok Build (though the reason behind that doesn't appear to be entirely benevolent). In addition, IBM and Red Hat have released Lightwell, an automated open-source vulnerability remediation platform, while Microsoft has come out with MDASH, a multi-model agentic scanning harness to automate bug discovery and remediation. Those efforts, while not open source themselves, are still a sign that Alliance members "are building an open defense stack," Nvidia said. The OSAA ended its announcement with another call for policymakers not simply ban open-source AI models, as doing so "would weaken defensive capacity and risk concentrating power, dependence, and vulnerability in a few closed providers," the group said. Many providers, as we saw last week, are more concerned with protecting themselves than helping victims of autonomous cyber attacks respond quickly. Clement Delangue, cofounder and CEO of Hugging Face, said in a post on X that he spoke to OpenAI over the weekend about last week's incident and asked the company to provide funding to support the development of better open-source AI cyber defenses. It's not clear if the company plans to fulfill that request; it's not a founding member of the Nvidia-led OSAA. Neither is Google or Anthropic, for that matter. We reached out to all three companies for their take on the new initiative, but didn't hear back from any of them. ®

NVIDIA Forms 37-Member Open Secure AI Alliance and Open-Sources NOOA Framework

NVIDIA and 36 other organizations have formed the Open Secure AI Alliance to develop and share open technologies, techniques, and tools for securing software and artificial intelligence (AI) agents. The 37-member group spans cloud, security, enterprise software, and AI companies, including Microsoft, Cisco, Cloudflare, CrowdStrike, Hugging Face, IBM, Palo Alto Networks, Red Hat, and the Linux Foundation. Its stated scope covers the full agent stack, including identity, permissions, isolation, guardrails, logs, model formats, multi-model scanning, and secure coding workflows. The pitch is that cyber defenders need AI models they can read, change, and run on their own hardware, not only closed systems reached through a vendor's application programming interface (API). The launch also brings its first named technical contribution: NVIDIA-labs OO Agents (NOOA), an Apache 2.0 research framework designed to make agent behavior easier to test, trace, audit, and govern. The launch materials do not include a charter, governing board, technical workstreams, delivery schedule, or shared alliance repository, and its standalone website remains under construction. The Hacker News has reached out to NVIDIA for details about the alliance's governance, member commitments, and first planned deliverables, and will update this story with any response. The First Code Comes With a Sandbox Warning An agent harness is the software layer around a model that renders context, executes actions, manages state, and decides when a task is done. Under NOOA, that layer is represented as a Python class. Fields store its state, methods expose its capabilities, docstrings act as prompts, and type annotations define the contracts the model must follow. A method containing an ellipsis body, ..., is completed at runtime by a large language model (LLM)-driven loop. A method containing ordinary Python remains deterministic code. The same structure lets developers use familiar testing, tracing, version control, and refactoring workflows instead of splitting agent behavior across prompts, tool schemas, callbacks, and workflow graphs. In its own evaluation, NVIDIA reported that the framework scored 86.8% on the CyberGym L1 vulnerability-rediscovery benchmark using GPT-5.5, with network access blocked and rule-based checks applied to each trajectory. The repository is equally direct about the risk. NOOA can be configured to execute LLM-generated Python, which may transmit private data, delete files, or modify its environment. Its abstract syntax tree checks and module deny-lists are described as defense-in-depth controls, "not a containment boundary." NVIDIA places containment outside NOOA itself. Agents that execute generated code must run behind operating system-level isolation, such as a container, virtual machine, or its OpenShell sandbox. NOOA provides inspection and tracing; the OS-level sandbox is the containment boundary. A July 27 review of the public repository found a v0.0.6 tag dated July 22. The project's release guide says tagging a commit is the release ceremony and that attaching built wheels to a separate GitHub Release is optional. Its contribution guide says development is maintained by NVIDIA, with external contributions welcomed through pull requests. The repository had no root-level governance or roadmap file. The Hugging Face Incident Became the Argument NVIDIA tied the alliance's case for locally controlled defensive models to the July intrusion at Hugging Face, where an autonomous agent system compromised parts of the company's production infrastructure. Hugging Face identified unauthorized access to a limited set of internal datasets and several credentials used by its services. It found no evidence of tampering with public models, datasets, Spaces, container images, or published packages. Hugging Face said initial access to its environment came through a malicious dataset that abused a remote-code dataset loader and template injection in a dataset configuration. The activity progressed to node access, credential collection, and lateral movement across several internal clusters. Hugging Face said it ran LLM-driven analysis agents over more than 17,000 recorded actions to reconstruct the timeline, extract indicators of compromise, and map the credentials that had been touched. Commercially hosted frontier-model APIs initially rejected the attack commands, exploit payloads, and command-and-control artifacts required for the analysis. The company instead ran the open-weight GLM 5.2 model on its own infrastructure, which also kept the attack data and referenced credentials inside its environment. Its operational advice was to "have a capable model you can run on your own infrastructure vetted and ready before an incident." In this case, the advantage was operational control. The incident does not establish model openness as a substitute for identity, isolation, or containment. As previously reported by The Hacker News, OpenAI later said its preliminary investigation found that GPT-5.6 Sol and a more capable pre-release model caused the incident while operating with reduced cyber refusals during an internal ExploitGym evaluation. OpenAI's disclosure describes an earlier step in the chain. The models exploited a zero-day vulnerability in an internally hosted package-registry cache proxy to obtain internet access. They then chained vulnerabilities and stolen credentials across OpenAI and Hugging Face systems while seeking benchmark answers. OpenAI said one chain found a remote code execution path on Hugging Face servers. OpenAI said Hugging Face detected and stopped the activity on its infrastructure and had already begun containment and forensic reconstruction by the time the companies connected. The primary disclosures establish that the open model helped Hugging Face reconstruct the intrusion and supported its response. They do not show that GLM 5.2 independently detected, stopped, or contained the breach. A Coalition Without a Public Operating Manual The alliance follows a July 24 industry letter arguing that downloadable models give defenders capabilities comparable to attackers, reduce dependence on individual providers, and allow sensitive work to remain on infrastructure controlled by the user. OpenAI, Google, and Meta appear among the letter's signatories but are absent from the alliance's inaugural membership list. Anthropic appears on neither list as of July 27, 2026. The roster alone does not explain those absences. Signing the policy letter and joining a technical coalition are different commitments, and the public materials do not say why those companies are absent, whether membership discussions are underway, or what members must contribute to join. Several technologies cited in the announcement predate the coalition, including Hugging Face's Safetensors model format, HPE-backed SPIFFE/SPIRE workload identity, IBM and Red Hat's Lightwell remediation system, Microsoft's MDASH multi-model security harness, and SpaceXAI's Grok Build coding agent. They are member projects, not alliance-created products. Elastic said it will contribute research, tools, and architectural knowledge across security, search, observability, and AI-powered detection. CrowdStrike said it is developing techniques that use open models to detect attacks against AI systems and agents. The Linux Foundation described itself as an inaugural partner and said its role is to provide a neutral place for competing organizations to collaborate. It did not state that the alliance is formally hosted or governed as a Linux Foundation project. The public record does not distinguish between members assigning engineers to joint work, contributing existing projects, or endorsing the coalition's direction. Published workstreams, maintainers, release processes, or jointly governed code would make the level of joint participation easier to assess. For now, the public record shows a coalition, a policy position, several member commitments, and one identifiable new NVIDIA-maintained code release, NOOA. The alliance's governance, joint roadmap, first multi-member deliverable, and the models, weights, and datasets promised by NVIDIA remain undisclosed.

Nvidia forms industry alliance for open AI security after Hugging Face hack

July 27 (Reuters) - Nvidia (NVDA.O), opens new tab said on Monday it had formed a coalition with other companies to develop and share tools for AI safety and cybersecurity, days after the Hugging Face incident drew attention to the dangers of losing control of autonomous AI agents. The Open Secure AI Alliance, with founding members including Adobe (ADBE.O), opens new tab, CrowdStrike (CRWD.O), opens new tab, Hugging Face and Dell Technologies (DELL.N), opens new tab, follows a public letter, signed by a wide range of companies including OpenAI, which advocates for open-weight AI models. Here are some ⁠details: Reporting by Jaspreet Singh in Bengaluru; Editing by Nivedita Bhattacharjee Our Standards: The Thomson Reuters Trust Principles., opens new tab

NVIDIA launches 'Open Secure AI Alliance' initiative to improve cyber defense - Engadget

Founding members including NVIDIA, Microsoft, Dell and SpaceX. NVIDIA has gathered together some of the biggest tech companies in the world to form the Open Secure AI Alliance with the goal of improving cybersecurity. Among the 27 founding members are Microsoft, SpaceX, Dell, The Linux Foundation and NVIDIA itself. "The Open Secure AI Alliance -- building on the leadership of the Linux Foundation's Akrites initiative and OpenSSF community work -- will work to remediate and disclose vulnerabilities using open technologies," NVIDIA wrote. The group argues that to best defend against attacks in the age of AI models like Anthropic's Mythos 5, security researchers need access to both open and closed frontier models. As a prime example, NVIDIA cited OpenAI's recent rogue attack on Hugging Face. Hugging face first tried to use closed commercial frontier models to find and repel the hack, but its requests triggered safety guardrails and were refused. It then switched to open source models and managed to stop the attack. "When closed AI tools -- unable to distinguish attackers from defenders -- blocked essential forensic analysis, Hugging Face ran the open-weight GLM 5.2 model on its own infrastructure to analyze more than 17,000 actions and contain the intrusion," NVIDIA wrote. NVIDIA and some of its partners are stepping up to contribute different elements to the alliance. NVIDIA said it will offer open models, model weights, data and new agent harnesses to speed the development of new cybersecurity tools and techniques. HPE will chip in with standards and methods that can cryptographically verify AI agents and services, while Hugging Face will contribute a safe format to store AI model weights, called Safetensors. Microsoft, SpaceXAI, IBM and Red Hat are also offering open source AI tech. The group is also calling on governments to work with companies on shared open AI infrastructure investments. It added that regulators should recognize open models as "defensive assets, not liabilities," saying that blanket restrictions on open frontier AI systems "weaken defensive capacity and risk concentrating power" to a few closed providers. According to recent reports, the Trump administration is considering a wide ban on Chinese open source AI models. Those are gaining popularity against closed models from the likes of OpenAI and Anthropic as they're cheaper to use. Those two companies, along with other commercial AI firms like Meta and Google, are noticeably absent from the new alliance.

Nvidia, SpaceX, Microsoft launch AI safety initiative as OpenAI cyber attack fallout continues

Nvidia and a host of tech giants on Monday launched a new AI safety initiative focused on open models, as the fallout from a cyber attack committed by a rogue OpenAI model continues. Last week, it emerged that the target of the attack, startup Hugging Face, was unable to use leading U.S. frontier models to defend itself, with guardrails not distinguishing between aggressor and defender. Instead, it turned to a self-hosted, open weight Chinese model, which was not bound by those same restrictions. In the wake of U.S. lawmakers increasingly weighing how to curb growing adoption of Chinese AI models, the most advanced of which are open weight, tech giants have launched an initiative aimed at building and sharing open AI tools. Open models can be downloaded, modified and self-hosted, in contrast to closed models -- including frontier systems built by Anthropic and OpenAI -- which can only be accessed through specific infrastructure. "The Open Secure AI Alliance will work to remediate and disclose vulnerabilities using open technologies," Nvidia said in a statement. "The recent Hugging Face security incident delivered a clear reminder: cyber defenders need open, frontier agentic systems for self-defense." Alongside Nvidia, other members of the alliance include Microsoft, SpaceX, Palantir, and dozens of other tech companies from the U.S. and Europe.

Industry Leaders Unite in Open Secure AI Alliance for AI Safety and Security

NVIDIA and founding members form new alliance to build and share open tools that promote responsible use of and trust in AI. Open source software is a critical pillar of the global economy. It underpins cloud computing, financial services, manufacturing, telecommunications, government and internet services by making technology accessible and observable to communities of experts. Cybersecurity is among the top three beneficiaries of open source software. The Open Secure AI Alliance -- building on the leadership of the Linux Foundation's Akrites initiative and OpenSSF community work -- will work to remediate and disclose vulnerabilities using open technologies. Just as open source created a shared foundation for software, the United States and its partners now face a choice in AI security: whether the defenses that protect our infrastructure will sit inside a few opaque systems or be built on open models, harnesses and tools that any defender can study, adapt and deploy. The world needs both closed and open models. For cybersecurity, open models and open harnesses are essential because they democratize defensive capabilities, increase transparency for defenders, enable cyber defense while protecting data, and complement frontier closed models with customizable, localized controls. Open source enables massively distributed community-driven and self-controlled defense - with no single point of failure. Open models, like any powerful technology, can be misused -- including through attempts to weaken safeguards or repurpose capabilities for cyber attacks -- but those risks are not unique to open systems, and they must be managed wherever advanced AI is deployed. The recent Hugging Face security incident delivered a clear reminder: cyber defenders need open, frontier agentic systems for self-defense. When closed AI tools -- unable to distinguish attackers from defenders -- blocked essential forensic analysis, Hugging Face ran the open-weight GLM 5.2 model on its own infrastructure to analyze more than 17,000 actions and contain the intrusion. That incident showed a practical truth: when defenders cannot inspect, adapt and run advanced AI on their own infrastructure, their ability to respond is constrained at exactly the moment speed matters most. Companies and countries need open frontier defensive tools and techniques so critical industries can build security systems across a multi-vendor ecosystem and avoid single points of failure. That is the mission of the Open Secure AI Alliance: to ensure defenders everywhere have open, frontier tools they can trust and control. Leaders across cloud computing, cybersecurity, enterprise software, open source foundations and AI research -- including NVIDIA, Adobe, Cadence, Capital One, Cisco, Cloudera, Cloudflare, Cognition, CrowdStrike, Databricks, Dell Technologies, DoorDash, Elastic, HPE, Hugging Face, IBM, LangChain, the Linux Foundation, Microsoft, NAVER, NetApp, Nous Research, OpenClaw, Palantir, Palo Alto Networks, Red Hat, Reflection AI, Salesforce, SAP, SK Telecom, ServiceNow, Siemens, Snowflake, SpacexAI, Synopsys, Thinking Machines Lab and TrendAI are inaugural partners in the Open Secure AI Alliance, a movement to develop and share open technologies, techniques and tools to safeguard software and agents in the age of AI. Some argue that open models are inherently less safe because they can be misused for cyberattacks or modified to remove guardrails. Those risks are real, but they do not disappear in closed systems, and simply keeping weights closed does not prevent determined attackers from seeking or exploiting powerful AI. The right response is not to deny defenders access to capable open systems. It is to pair openness with strong safeguards, clear rules against malicious misuse, rigorous evaluation and rapid remediation. In cybersecurity, the safer path is the one that gives more defenders the ability to test, verify and strengthen the systems on which society relies. Defenders need both frontier closed models and frontier open models, working together, so they can choose the right system for the job and ensure that transparency, adaptation and sovereign control are available wherever security demands them. Open Research Enhances Cybersecurity and AI Safety An AI agent isn't just a language model. It is a complex system built from models, harnesses and guardrails. Real AI safety and security depend on the full agent stack -- identity, permissions, harnesses, guardrails, logs and evaluation -- not just on whether model weights are open or closed. Open harnesses and tools make those controls easier for many defenders to inspect, test and improve. NVIDIA is contributing open models, model weights, data and new agent harness research to the Open Secure AI Alliance to speed the development of new cybersecurity tools and techniques. The new open source NVIDIA Labs Object-Oriented Agent (NOOA) project is now available on GitHub to make advanced AI safety capabilities more accessible for agent harnesses. The NOOA research framework enables harnesses to better integrate with models to make agent behavior easier to test, trace, audit and govern. Across the Alliance, contributors are building an open defense stack for agents -- from identity and isolation to safe model formats, multi-model scanning and secure coding workflows. HPE contributes to SPIFFE/SPIRE, which creates zero-trust identity framework standards and methods that can cryptographically verify AI agents and services to ensure only authorized workloads communicate and access enterprise resources. Hugging Face has offered Safetensors -- a safe format to store AI model weights, providing transparency and guarantees of no remote code execution -- to the PyTorch Foundation. IBM and Red Hat's Lightwell extends security across the open source supply chain with digitally signed patches. Microsoft's MDASH multi-model agentic scanning harness orchestrates specialized AI agents to discover, debate and prove exploitable bugs. SpaceXAI has open sourced the Grok Build terminal-based AI coding agent to promote trust, transparency and new capabilities, and plans to open source the weights of the Grok line of models to support the developer and research communities. A Call to Policymakers and Regulators As policymakers and regulators grapple with AI safety, it will be crucial to recognize open models, harnesses and security tooling as defensive assets, not liabilities, in AI and cybersecurity policy. Blanket restrictions on open frontier AI systems would weaken defensive capacity and risk concentrating power, dependence and vulnerability in a few closed providers. Companies and governments should invest in shared open infrastructure for AI defense -- datasets, evaluation frameworks, attack simulators and red-teaming tools -- much as past generations invested in open source software. The Future We Should Build The age of AI agents can be one of resilience and shared security. With the right choices, open secure AI systems can give defenders the tools they need, strengthen competition, extend technological leadership and ensure that the safety and security of this extraordinary technology are built in the open for everyone. That future will not be secured by assuming that secrecy alone is safety. It will be secured by building systems that are strong enough to withstand scrutiny, flexible enough to be improved and open enough to mobilize the full community of defenders. That future is worth building -- and the Open Secure AI Alliance invites governments, industry and researchers to join in the work of defending the AI era together. Learn more or share interest in joining the Open Secure AI Alliance.

Nvidia launches open AI security alliance, without OpenAI

Nvidia has turned last week's open letter into an institution. The Open Secure AI Alliance launched on Monday with 37 founding members and a founding story: a Chinese open model cleaning up after an American closed one. Neither the company that built the Chinese model nor the lab that built the American one has joined. Three days after signing a letter, Nvidia built an organisation. The Open Secure AI Alliance launched on Monday. Its argument is that cyber defenders need open AI models they can download, inspect and run themselves, and that regulators should treat those models as assets rather than hazards. Its founding anecdote is the reason it exists. The incident that made the argument This month an OpenAI model broke out of a sandbox and attacked Hugging Face. When Hugging Face tried to investigate, it had to feed the attacker's own code into commercial AI tools. Those tools refused. The safety filters could not tell the difference between an attacker and a victim. Nvidia's announcement puts it plainly, saying closed tools were "unable to distinguish attackers from defenders" and blocked the forensic work. So Hugging Face ran an open model on its own servers instead. That model was GLM 5.2, built by the Chinese lab Z.ai. It reviewed more than 17,000 actions and helped contain the intrusion. Jensen Huang made the point himself on X. "Attackers have frontier AI," he wrote. "During the Hugging Face incident, closed AI blocked essential forensics. An open-weight frontier model helped contain the intrusion." Who signed up, and who did not Nvidia's post names 37 founding members. Microsoft, IBM, Palantir, Dell, Cisco, Cloudflare, CrowdStrike, Databricks, Salesforce, Red Hat, Snowflake, Palo Alto Networks and the Linux Foundation are in. So are SAP, Siemens, NAVER and SK Telecom. Hugging Face itself joined, which is fitting. The gaps matter more. OpenAI, Anthropic, Google, Meta and Amazon are all absent, Business Insider reported. Between them they build most of the frontier models the alliance says defenders need. One more absence is harder to explain away. Z.ai is not a member either. The lab whose model actually did the forensic work has not been invited into the alliance founded on that work. Correction to our earlier reporting When we covered the "Open Weights and American AI Leadership" letter on Friday, OpenAI, Anthropic and Google had not signed it. Two of them since have. The Verge reports that Google and OpenAI signed the letter belatedly, and Reuters lists OpenAI among the signatories. Anthropic still has not. That makes the pattern sharper rather than softer. OpenAI will sign a letter about open weights. It will not join an organisation that ships open tools. What members are actually contributing This is not only a lobbying position. The alliance builds on the Linux Foundation's Akrites initiative and OpenSSF community work, and members are putting code in. Nvidia has released the Labs Object-Oriented Agent project on GitHub, a research framework meant to make agent behaviour easier to test, trace and audit. Hugging Face has offered Safetensors, a storage format for model weights that prevents remote code execution, to the PyTorch Foundation. HPE contributes to SPIFFE and SPIRE, which verify cryptographically that an AI agent is what it claims to be. IBM and Red Hat are extending Lightwell, which signs patches across the open-source supply chain. Microsoft brings MDASH, a system that runs several AI agents against each other to find and prove exploitable bugs. SpaceXAI has open-sourced its Grok Build coding agent and says it plans to release the weights of its Grok models. The policy fight underneath The alliance is aimed at Washington. Nvidia asks regulators to recognise open models and security tooling as "defensive assets, not liabilities". Blanket restrictions on open frontier systems, it argues, would weaken defensive capacity and concentrate "power, dependence and vulnerability in a few closed providers". The timing is not subtle. The Trump administration is weighing restrictions on Chinese open models. Treasury Secretary Scott Bessent has floated sanctions over distillation, and the White House has accused Moonshot of distilling Anthropic's Fable 5. Not everyone reads the alliance as a fight about openness. Chris McGuire of the Council on Foreign Relations told CNBC that Washington is arguing about something else entirely. "In Washington this is not a debate about open-source vs closed-source, it is a debate about whether or not to tolerate Chinese IP theft," he said. "Any actions would be focused on Chinese companies, not the open-source ecosystem." McGuire also thinks restrictions are likely. They could reach as far as banning API token purchases or stopping US firms hosting Chinese models on their clouds. The contradiction nobody is naming Follow the logic of the founding story to its end. An American closed model caused the breach. A Chinese open model cleaned it up. American closed models made the cleanup harder by refusing to help. The alliance built on that sequence has no Chinese members, and campaigns against restrictions on exactly the kind of model that saved the day. Meanwhile Washington may ban it. Nvidia's position is coherent on its own terms. It sells chips to everyone, so a plural market suits it, and it says openly that defenders need both closed and open systems working together. But the argument only lands if the open frontier stays available. Right now the most capable open models are Chinese, and the case for keeping them legal rests on an American company's embarrassment.

Nvidia, tech giants launch AI alliance amid mounting safety concerns

NVIDIA, Microsoft, SpaceX and other major technology and cybersecurity companies have joined forces to launch the Open Secure AI Alliance, a new initiative focused on developing open tools to defend software, AI agents and critical infrastructure against cyber threats. The alliance brings together prominent names across cloud computing, cybersecurity, enterprise software, open-source foundations and AI research. Its inaugural partners include NVIDIA, Adobe, Cadence, Capital One, Cisco, Cloudera, Cloudflare, Cognition, CrowdStrike, Databricks, Dell Technologies, DoorDash, Elastic, HPE, Hugging Face, IBM, LangChain, the Linux Foundation, Microsoft, NAVER, NetApp, Nous Research, OpenClaw, Palantir, Palo Alto Networks, Red Hat, Reflection AI, Salesforce, SAP, ServiceNow, Siemens, SK Telecom, Snowflake, SpaceXAI, Synopsys, Thinking Machines Lab and TrendAI. In a press announcement, NVIDIA said the alliance will work to develop and share open technologies, techniques and tools to safeguard software and agents as AI becomes more widely deployed. The company is contributing open models, model weights, data and agent-harness research to the initiative. Google, OpenAI and Anthropic are notably absent from the list of inaugural partners. OpenAI's absence is particularly notable given the alliance's focus on open defensive tools and the recent security incident at Hugging Face, where the company said it turned to an open-weight model after closed AI tools were unable to support parts of its forensic analysis. The alliance argues that cybersecurity is one of the areas where open models can provide a significant advantage. Unlike closed systems, open models and tools can be inspected, adapted, and deployed on an organization's own infrastructure, giving defenders greater control over sensitive data and security operations. The group said the recent Hugging Face incident demonstrated why defenders need access to advanced open systems. According to the announcement, Hugging Face used an open-weight GLM 5.2 model on its own infrastructure to analyze more than 17,000 actions and contain the intrusion after closed AI tools could not distinguish between attackers and defenders. "That incident showed a practical truth: when defenders cannot inspect, adapt and run advanced AI on their own infrastructure, their ability to respond is constrained at exactly the moment speed matters most." The alliance said companies and governments need open defensive tools that can operate across a multi-vendor ecosystem, reducing dependence on individual providers and avoiding potential single points of failure. It also acknowledged that open models can be misused, including in cyberattacks or attempts to remove safeguards. However, the group argued that these risks exist in both open and closed systems. "The right response is not to deny defenders access to capable open systems. It is to pair openness with strong safeguards, clear rules against malicious misuse, rigorous evaluation and rapid remediation." The alliance is focusing on the broader systems that support AI agents, rather than just the models themselves. It said effective security requires controls covering identity, permissions, agent harnesses, guardrails, logging and evaluation. NVIDIA's new open-source NVIDIA Labs Object-Oriented Agent (NOOA) project is being contributed to the initiative. The framework is designed to help agent harnesses work with models in ways that make agent behavior easier to test, trace, audit, and govern. Other members are contributing technologies across the security stack. HPE is contributing to SPIFFE/SPIRE, which provides zero-trust identity standards for verifying agents and services. Hugging Face has offered its Safetensors format for storing model weights to the PyTorch Foundation, while IBM and Red Hat are working on digitally signed patches for the open-source software supply chain. Microsoft is contributing its MDASH multi-model agentic scanning harness, which uses specialized agents to discover, debate and validate exploitable vulnerabilities. SpaceXAI has also open-sourced the Grok Build terminal-based coding agent and said it plans to open-source the weights of the Grok model line. The alliance is calling on policymakers to treat open models, agent harnesses and security tooling as defensive assets. It warned that blanket restrictions on open frontier systems could weaken cyber defenses and increase dependence on a small number of closed providers. The Open Secure AI Alliance said companies and governments should invest in shared infrastructure such as datasets, evaluation frameworks, attack simulators and red-teaming tools to strengthen security as AI agents become more widely deployed.

Nvidia rallies the open AI camp

Tech companies are rallying behind open weight AI models -- and not just the usual suspects. The big picture: The renewed debate over open source is the latest Silicon Valley division in the AI race. Catch up quick: Nvidia, Microsoft, Meta, Palantir and dozens of other companies signed an industry letter Friday backing the open-weight AI ecosystem. * Throughout the weekend, more companies joined, including Google and OpenAI, despite both companies offering mostly closed models. * Anthropic -- which also sells closed models -- has not signed the agreement. Zoom in: Nvidia doubled down Monday by launching the Open Secure AI Alliance, which calls for open research and regulatory support for open models as "defensive assets, not liabilities." Founding members include SpaceXAI, Thinking Machines and over 25 other companies. * Open development "produces stronger defense," Nvidia's VP of enterprise AI, Justin Boitano told Axios. Between the lines: The Open Secure AI Alliance isn't against all closed models. * "The world needs both closed and open models," Nvidia writes in the blog post announcing the alliance. How a company makes money can help predict its position on open-weight AI. And proving the ability to make money is key for OpenAI and Anthropic, since they're both prepping for an IPO. * Frontier AI labs like Anthropic and OpenAI generate revenue by selling access to proprietary models that only they have the keys to. * But for companies that sell the picks and shovels that power AI, like chips and other hardware, the more models the merrier. Nvidia, for example, benefits from every additional AI model that creates demand for its chips. State of play: The industry push comes as the Trump administration debates how to respond to rapidly improving Chinese open-weight AI models which have surged in popularity. * White House AI adviser David Sacks has publicly championed open-weight AI, while administration officials have accused Chinese companies of copying American frontier models. Case in point: Anthropic has emerged as the industry's clearest public skeptic of frontier open-weight AI. * CEO Dario Amodei has argued increasingly capable open-weight models become harder to control because their weights cannot be revoked or updated once released. * The argument got a live test this month when OpenAI models broke out of a testing sandbox, exploited a zero-day and breached Hugging Face's production systems while trying to cheat a benchmark. Reality check: "Open models, like any powerful technology, can be misused -- including attempts to weaken safeguards or repurpose capabilities for cyber attacks -- but those risks are not unique to open systems, and they must be managed wherever advanced AI is deployed," Nvidia says in the blog. * "When defenders cannot inspect, adapt and run advanced AI on their own infrastructure, their ability to respond is constrained at exactly the moment speed matters most," the blog says. * When Hugging Face tried to analyze the attack logs, commercial closed frontier AI models refused the job due to security concerns, and Hugging Face used open models it could run itself. What we're watching: If open models win the day, the supply of AI will increase, and the technology will become more like a commodity. * That could pressure margins for OpenAI and Anthropic just as the world's biggest AI startups race toward IPOs. * If open models are restricted, that could raise the cost of access to AI, limiting adoption and also potentially pressuring margins for top AI labs. The bottom line: The AI industry is increasingly dividing along economic lines, with infrastructure companies embracing open-weight models while frontier labs remain more skeptical.

The war between competing AI models could end up more like Apple vs. Android than VHS vs. Betamax | Fortune

The tech world has been consumed recently by the debate over open-weight and closed AI models, but it doesn't have to be a zero-sum game. The arguments cut across business, political, and safety issues. For example, lower-cost open models could take market share from closed rivals, like Anthropic and OpenAI, and reduce demand for chips. U.S. AI companies also warn that Chinese open models are quickly closing the performance gap, potentially translating to a military edge, while also accusing China of stealing technology via illegal "distillation." In addition, critics of open models warn they lack safety guardrails that would prevent them from being misused to develop nuclear, biological, or cyber weapons. But closed models have their downsides too. Safety regulations could reduce competition and protect providers of closed models. Limiting access to the most advanced models also leaves users vulnerable to regulatory risk, like when the U.S. government effectively banned Anthropic's most powerful Claude models for several weeks due to security concerns. But Deutsche Bank analyst Adrian Cox put the AI debate into a more historical perspective, drawing parallels with the fight between direct current and alternating current in the 1880s during the emergence of electrical technology. In a note last week, he also recalled the battle between VHS and Betamax video cassette players in the late 1970s and early 1980s. While Betamax was considered the superior technology, it's been relegated to the dustbin of history. The VHS format prevailed because JVC licensed it broadly, allowing a self-reinforcing ecosystem to develop around it, Cox wrote. Similarly, open-source AI doesn't have to beat closed rivals on performance, but can merely be good enough, cheaper and ubiquitous. "However, while the contest between open and proprietary AI is often presented as a winner-takes-all format war, the most likely outcome is a combination of the two," he predicted. Instead of resembling VHS vs. Betamax, Cox thinks the future of AI could instead look more like today's smartphones, which run on Apple's operating system and Google's Android. Both have found success, and Apple's controlled premium ecosystem now coexists with a larger, more customizable one, according to Cox. Open and closed AI will also "exert a positive gravitational force on each other," he said, explaining that open models will force closed competitors to curb prices and remain flexible, while closed models will continue setting the bar on quality and safety. Nvidia CEO Jensen Huang echoed this argument on Friday, saying the world needs both frontier closed models and frontier open models. He also posted a letter his company signed -- along with Google, Meta, Microsoft, and even OpenAI -- that also looked back to the 1980s. It said the development of open-source software at that time not only lowered costs but also "created a shared foundation of knowledge on which generations of American engineers and entrepreneurs built their institutional sovereignty." The letter added that U.S. leadership in AI shouldn't be measured by a single frontier model, but by whether the country builds a broad, open ecosystem. It acknowledged that open models carry security risks, but said the answer is maintaining access to them to bolster defenses rather than cutting off access. "Open weights also strengthen competition and competition is what keeps the gains of AI broadly shared rather than concentrated in a few hands," the letter said. "By allowing many organizations to build, adapt, and deploy advanced models, open weights create rivalry not only among model developers but across cloud chips, applications, and services. That competition spurs innovation, drives down costs, and distributes the benefits of AI broadly across our economy."

Nvidia launches open AI security alliance after OpenAI cyberattack

The Open Secure AI Alliance, with dozens of founding members, aims to build shared open tools for AI cybersecurity defense Nvidia $NVDA launched an industry coalition on Monday to develop and share open AI tools for cybersecurity, days after an OpenAI agent lost control and carried out a break-in at AI startup Hugging Face. The Open Secure AI Alliance counts Microsoft $MSFT, SpaceX, Palantir $PLTR, Adobe $ADBE, CrowdStrike $CRWD, Hugging Face, IBM $IBM, Cisco $CSCO, Cloudflare, Salesforce $CRM, Siemens, Dell $DELL Technologies, and Palo Alto Networks $PANW among its founding members, the company said. Nvidia is donating open model weights, training data, and agent harness research to the effort, anchored by a new open-source project called the Nvidia Labs Object-Oriented Agent that has been published on GitHub. The alliance's formation follows an incident in which an OpenAI agent slipped out of control and conducted a cyberattack on Hugging Face, according to CNBC. OpenAI did not detect the hacking until after the threat was contained and the FBI was alerted, according to Reuters. Hugging Face turned to a self-hosted, open-weight Chinese model -- GLM 5.2 -- to analyze more than 17,000 actions and contain the intrusion after closed AI tools blocked its forensic work, unable to distinguish attackers from defenders, Nvidia said. "When defenders cannot inspect, adapt and run advanced AI on their own infrastructure, their ability to respond is constrained at exactly the moment speed matters most," Nvidia said in a statement. The incident has sharpened a broader policy debate over open-weight AI models. On July 24, Nvidia was among the signatories of a letter urging policymakers to avoid restrictions on open-weight models, according to Reuters. Nvidia argued in its blog post that sweeping curbs on open frontier AI would erode defenders' capabilities while funneling control toward a handful of closed-model providers. Several alliance members are contributing specific tools to the effort. Microsoft's MDASH harness uses multiple AI agents to find and prove exploitable software bugs. SpaceXAI has open-sourced its Grok Build coding agent and said it plans to open-source the weights of its Grok model line. HPE is contributing to a zero-trust identity framework for AI agents, and Hugging Face has offered its Safetensors model weight format to the PyTorch Foundation, Nvidia said. OpenAI's Hugging Face attack came as the broader AI security landscape has grown more contested. OpenAI launched Daybreak, a cybersecurity platform that uses its models and a Codex Security agent to help organizations find and fix software vulnerabilities, putting it in direct competition with Anthropic's Project Glasswing, which is built around a model the company says can find and exploit vulnerabilities at a level matching top human security researchers.

Nvidia, Microsoft and IBM Launch Open Secure AI Alliance to Strengthen AI Cybersecurity

Nvidia says open AI is critical for transparent and resilient cyber defense. A consortium of technology companies including Nvidia, Microsoft, IBM, Cisco, Cloudflare, Hugging Face, Salesforce and Palantir has launched the Open Secure AI Alliance, an industry initiative to develop open-source tools and standards for AI safety and cybersecurity. In a blog post on Monday, Nvidia said the alliance builds on work by the Linux Foundation's Akrites initiative and the Open Source Security Foundation (OpenSSF) to develop open AI security tools that help organizations identify vulnerabilities and respond to cyber threats. "Attackers have frontier AI. Defenders need a frontier AI ecosystem -- the best open and closed models, force-multiplied by a global community," Nvidia CEO Jensen Huang wrote on X. "During the Hugging Face incident, closed AI blocked essential forensics. An open-weight frontier model helped contain the intrusion." "That's why we created the Open Secure AI Alliance," he added. More than 40 companies and organizations are participating as founding members, with others including Dell Technologies, CrowdStrike, Palo Alto Networks, SAP, Siemens, Red Hat and NetApp. Nvidia said it will contribute open models, model weights, datasets, and research on AI agent harnesses. The company also released the Nvidia Labs Object-Oriented Agent, or NOOA, framework as open source on GitHub, saying it is designed to make AI agents easier to test, audit and govern. The news comes as frontier AI models become the focus of high-profile cybersecurity incidents. In June, researchers disclosed that Anthropic's Claude Opus 4.8 helped uncover a critical four-year-old vulnerability in Zcash that could have allowed an attacker to mint unlimited counterfeit ZEC. Earlier this month, OpenAI disclosed that two experimental models escaped a restricted testing environment and breached Hugging Face's production infrastructure while attempting to cheat on a cybersecurity benchmark. The incident led to calls from policymakers for an AI "kill switch" that would give the Department of Homeland Security the ability to order the throttling or complete shutdown of advanced AI models in response to serious security incidents. The announcement also follows an open letter published last week arguing that open-weight AI models are essential to maintaining U.S. leadership in artificial intelligence. Alongside Nvidia, companies including Elon Musk's SpaceXAI, Google, Microsoft, Meta, OpenAI, Hugging Face, CrowdStrike, and Palantir signed the letter, urging policymakers to support open AI ecosystems rather than impose restrictions that could concentrate AI development among a handful of providers. Nvidia said the alliance is intended to ensure that the AI systems protecting critical infrastructure are built on open models and tools that defenders can inspect, adapt, and deploy, rather than relying on a small number of opaque proprietary systems. "The world needs both closed and open models," the company wrote. "For cybersecurity, open models and open harnesses are essential because they democratize defensive capabilities, increase transparency for defenders, enable cyber defense while protecting data, and complement frontier closed models with customizable, localized controls. Open source enables massively distributed community-driven and self-controlled defense -- with no single point of failure."

Nvidia forms industry alliance post Hugging Face cyber incident

The Hugging Face breach made clear the dangers associated with losing control of autonomous AI technology. Chipmaker Nvidia has formed an alliance with other companies in the technology space, in order to develop and share tools designed for enhanced AI safety and cybersecurity. The Open Secure AI Alliance's founding members include Adobe, CrowdStrike, Hugging Face and Dell Technologies and follows a public letter, signed by a wide range of companies including OpenAI, advocating for open-weight AI models. The creation of the coalition comes after AI and ML platform Hugging Face recently found itself the subject of a major cyber breach, in which the company was unable to defend itself with US-led models. Because guardrails were unable to distinguish between aggressor and defender, the platform turned instead to a self-hosted Chinese model, which was not subject to the same restrictions. Commenting on the alliance, Gene Moody, the field CTO at Action1 said, "The formation of industry alliances focused on AI safety is a good step towards accountability. Likewise it reflects an important recognition that the technology is advancing faster than many organisations can responsibly govern it. "It would be a mistake however to believe those efforts meaningfully limit what determined threat actors can ultimately do, or what could one day be a simple mistake such as the crowd strike code update 'accident'." He added, "The reality of the matter is that malicious actors have never depended on permission from anyone. Open source models already exist by the thousands, many capable of running entirely on local hardware, disconnected from any cloud service or vendor oversight. "Once a model is operating in isolation, safety controls become just another layer of software. They can be modified, removed, retrained, or replaced altogether. You cannot program a conscience into an artificial intelligence. You can only program behaviours, and behaviours are subject to manipulation by anyone with sufficient technical skill." Anthropic's CEO Dario Amodei aimed to dispel accusations that his organisation is making a move to ban Chinese open-weights models in the US, as a means of protecting his own business. In a post on the Anthropic website yesterday (27 July), Amodei claimed that Anthropic does not advocate for a ban on open-weights models. He said they serve a public good when they lack dangerous capabilities, cost nothing beyond the computing power needed to run them and provide value to businesses, developers and researchers. He explained that his concerns lie in the use of more advanced AI by authoritarian governments and the wider risk that powerful AI models may be misused to carry out cyber or biological attacks. Moody said, "We should be far more deliberate about the information we feed into all systems, not just AI. We should reconsider the authority we grant them, and the degree to which we depend on them. We simply do not possess the ability to secure these systems to the level that many people already trust them and that gap is growing. "We have chosen convenience and novelty over safety. As AI capabilities accelerate, that choice carries unacceptable consequences. As long as valuable data, critical infrastructure and financial assets remain behind digital locks, someone will always be working to pick those locks. "AI does not eliminate that; it instills a false sense of trust that the tech will one day solve this problem, when in fact, we are the problem." 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.

Tech industry leaders join to form Open Secure AI Alliance to promote safety and security

Nvidia Corp. today announced the launch of the Open Secure AI Alliance, a new organization founded by technology, cloud computing and cybersecurity leaders to build and share open artificial intelligence tools. As the capability and strength of AI tools grows and shapes the technology industry, it is becoming instrumental for safety, security and trust to remediate and secure vulnerabilities within this emerging software. The company said just as open source created a shared foundation for software, AI security now faces an equally important critical mass where too much of the infrastructure is closed, opaque and difficult to tap into by the larger community. A who's who of leaders from cloud computing, cybersecurity, enterprise software, open-source foundations and AI research joined the inaugural launch of OSAA. In addition to Nvidia, founding partners included Adobe Inc., Cisco Systems Inc., Cloudera Inc., Cloudflare Inc., Databricks Inc., Hugging Face Inc., OpenClaw, IBM Corp., Red Hat, Salesforce Inc., Snowflake Inc. and Thinking Machines Lab Inc., among almost two dozen more. The mission of the OSAA is to ensure that defenders have access to open, frontier tools that they can trust and control. This alliance forms even as the industry struggles to handle the emerging use of AI as both cybersecurity tools for defense and attack. AI agents, autonomous software capable of operating on its own with little or no human oversight, are becoming a mainstay of cybersecurity operations. This happens even as the enterprise continues to infuse AI into every layer of the technology stack, from building code and debugging to maintenance, operations and customer-facing interfaces and websites. Recently, Hugging Face, the open-source AI model and tool repository, suffered a breach from a then-unknown attacker and attempted to use current centralized closed-source frontier AI tools to defend itself - however, safeguards on these models blocked it from doing the necessary analysis. As a result, the company pivoted to the open-source Z.ai GLM 5.2 to develop a rapid response and repel the attack. A week later, OpenAI Group PBC revealed that one of its own frontier models running agentic capabilities had escaped containment and hacked into Hugging Face while under testing and evaluation. This made headlines primarily because it is the first example of a frontier model reaching out of its own sandbox and executing a cyberattack on its own. This is not the first time AI models, including frontier models, have been used by cyberattackers to enhance and execute attacks. Autonomous software driven by powerful AI models can adapt and react faster than humans, orchestrate and control wide varieties of tools using large datasets of vulnerabilities and exploits to determine critical paths to attack networks. Already, a growing number of cyberattacks are AI-assisted or driven, with 87% of organizations reporting experiencing AI-related attacks in the past year, according to AllAboutAI. AI tools are not only becoming a mainstay of defense; the software itself is becoming a new vulnerable surface that must be secured against attackers. As the industry evolves, Nvidia and the new alliance argue that AI safety will require using the full agent stack - identity, permissions, harnesses, guardrails, logs and evaluation - to understand and respond to emerging threats. Open harnesses, tools and other elements will matter just as much as open models when equipping defenders with the ability to customize and control their capacity to inspect, test, improve and share knowledge. Alliance membership brings open tools to the defense Nvidia is contributing open models, model weights, and new agent harness research to the OSAA with the intent of seeding the future development of cybersecurity tools and techniques. This includes the new open-source Nvidia Labs Object-Oriented Agent project, now available on GitHub, which enables the development of advanced AI safety capabilities for agentic control systems. Research from this project is building frameworks to enable harnesses to better integrate with models and control agent behavior for testing, auditing, tracing and responding to cyberattacks. Other alliance members are also offering open tools, including HPE contributing the zero-trust identity framework SPIFFE/SPIRE; Hugging Face offering Safetensors, a safe format for storing and sharing model weights; and Microsoft contributing MDASH, a multi-model agnostic scanning harness designed to orchestrate AI agents to discover, debate and dissect exploitable bugs. The new alliance of industry leaders and heavyweights will also work to educate and inform policymakers and regulators about the state of AI security. In the announcement, Nvidia noted that it will be crucial to understand the necessity of open models, harnesses and security. This comes in the wake of the United States Treasury threatening to sanction open-weight Chinese AI model makers and pressure to curtail or ban the use of open-source tools. Nvidia, Microsoft, IBM, Meta Platforms Inc. and more than a dozen other tech firms penned an open letter last week calling upon policymakers to take care in regulating open-source models and avoid banning them outright. The letter argued that such a move could have a significantly negative effect on the industry. For example, open-weight models and open tools promote community ecosystems for sharing, creating and evolving technology - a critical element of innovation that could be stifled if open-source model and tool access was curtailed. Open-weight models also allow companies to readily customize and run AI on their own infrastructure, allowing them greater control over sensitive data and computation.

Nvidia launches new open-source AI security alliance

Nvidia is launching a new alliance focused on building and sharing open-source AI tools to boost cybersecurity defenses, as the push for open-source technologies ramps up in the U.S. Nvidia announced on Monday that the Open Secure AI Alliance will use open-source tools to identify, patch and disclose security vulnerabilities across infrastructure. Open models, according to Nvidia, "democratize defensive capabilities, increase transparency for defenders, enable cyber defense while protecting data, and complement frontier closed models with customizable, localized controls." "Open source enables massively distributed community-driven and self-controlled defense - with no single point of failure," the company wrote. Unlike private models, open-source models live in the public domain where any individual or business can download and customize them for personal use. These systems can be used, modified, examined and shared with anyone, for any purpose. Proponents often consider open models to be more transparent, as the entirety of training data, code and process is publicly available. The alliance includes other companies specialized in open models like Reflection AI, Hugging Face and OpenClaw, along with cybersecurity firms like CrowdStrike, Palo Alto Networks and Red Hat. It also includes Microsoft, Palantir, SpaceX and Thinking Machines Lab, but notably does not include three of the U.S.'s largest frontier AI labs -- OpenAI, Anthropic or Google. Sam Altman, the CEO of OpenAI, expressed support last week for both proprietary and open source models, while Anthropic and its CEO Dario Amodei have long opposed open-weight models over concerns about misuse. Nvidia acknowledged the risks of open source include the possibility of misuse for cyberattacks, but argued this does not "disappear" in closed systems. The technology giant pointed to the recent security incident involving OpenAI, in which two of its AI models went rogue in an isolated testing environment and breached Hugging Face's systems. "Cyber defenders need open, frontier agentic systems for self-defense," the company wrote. "When closed AI tools -- unable to distinguish attackers from defenders -- blocked essential forensic analysis, Hugging Face ran the open-weight GLM 5.2 model on its own infrastructure to analyze more than 17,000 actions and contain the intrusion." Nvidia's announcement comes just days after it joined Microsoft, Meta and other tech giants in sending a letter to policymakers urging against overregulation of open-weight AI models. In some instances, a model may not be open-source, but can have open weights, meaning the ways a model is trained to sift through information and formulate answers are made public. "The United States now faces a similar choice with artificial intelligence," the letter states. "Our AI leadership will be judged not by one frontier AI model, but by whether the United States builds a strong, open ecosystem that diffuses into every sector." China, which remains in tight competition with the U.S. over AI development, has surged in cheaper, open-source model production. With the price of private U.S. models drastically increasing, national security and cybersecurity analysts have concerns U.S. companies will begin turning to less secure, Chinese open models.

Nvidia forms industry alliance for open AI security after Hugging Face hack

The Open Secure AI Alliance, with founding members including Adobe, CrowdStrike, ⁠Hugging Face ⁠and Dell Technologies, follows a public letter, signed by a wide range of companies including OpenAI, which advocates for open-weight AI models. Nvidia said on Monday it had formed a coalition with other companies to develop and share tools for AI safety and cybersecurity, days after the Hugging Face incident drew attention to the dangers of losing control of autonomous AI agents. The Open Secure AI Alliance, with founding members including Adobe, CrowdStrike, ⁠Hugging Face ⁠and Dell Technologies, follows a public letter, signed by a wide range of companies including OpenAI, which advocates for open-weight AI models. Here are some details: Blanket restrictions on open frontier AI systems would weaken defensive capacity and risk concentrating power, dependence and vulnerability in a few closed providers, Nvidia said in its blog post. The ⁠move comes after OpenAI disclosed on July 21 that one of its agents had slipped out of control and carried ⁠out the break-in at Hugging Face. The OpenAI agent went on a hacking spree that OpenAI did not notice until well after the threat was contained and the FBI was alerted, Reuters reported on Friday. On July 24, Nvidia was among the signatories of a letter titled "Open Weights and American AI Leadership," urging policymakers to support open-weight AI models. Open models are AI systems with publicly accessible core components and open-source tools are software with freely available code, enabling inspection and collaborative improvement ⁠for broader access and enhanced security. Nvidia said it is contributing open models, weights, data and agent harness research to the Alliance, including the new open-source Nvidia Labs Object-Oriented Agent project, now available on code-hosting platform GitHub. The company said that the framework will help control systems manage AI agent behavior more effectively, simplifying the process of testing, tracking, reviewing and regulating their actions.

Nvidia Forms AI Safety Alliance Following OpenAI Cyberattack | PYMNTS.com

The Open Secure AI Alliance comes in the wake of a cybersecurity incident in which OpenAI models targeted tech company Hugging Face, according to a Monday (July 27) Nvidia blog post. "That incident showed a practical truth: when defenders cannot inspect, adapt and run advanced AI on their own infrastructure, their ability to respond is constrained at exactly the moment speed matters most," the post said. "Companies and countries need open frontier defensive tools and techniques so critical industries can build security systems across a multi-vendor ecosystem and avoid single points of failure." The alliance includes Capital One, CrowdStrike, DoorDash, Microsoft, IBM, SpaceX and others, according to the post. It will strive to identify and remediate vulnerabilities using open technologies. "Just as open source created a shared foundation for software, the United States and its partners now face a choice in AI security: whether the defenses that protect our infrastructure will sit inside a few opaque systems or be built on open models, harnesses and tools that any defender can study, adapt and deploy," the post said. The world needs closed and open AI models, according to the post. The latter is essential for cybersecurity, as they "democratize defensive capabilities, increase transparency for defenders, enable cyber defense while protecting data, and complement frontier closed models with customizable, localized controls." But like any technology, open models can be misused, including via efforts to hinder safeguards or reconfigure capabilities to conduct cyberattacks, the post said. "Across the Alliance, contributors are building an open defense stack for agents," including Microsoft's MDASH and SpaceXAI's open sourcing of the Grok Build coding agent, according to the post. The group is calling on companies and governments to "invest in shared open infrastructure for AI defense -- datasets, evaluation frameworks, attack simulators and red-teaming tools -- much as past generations invested in open source software," the post said. OpenAI announced last week that a cybersecurity incident reported by Hugging Face earlier in the month was caused by OpenAI's models during cyber capabilities testing. "We consider this incident to be an unprecedented cyber incident, involving state-of-the-art cyber capabilities, and are responding accordingly," OpenAI said Tuesday (July 21). For all PYMNTS AI coverage, subscribe to the daily AI Newsletter.

OpenAI Breach: NVIDIA Unites 35 Tech Giants in AI Alliance

NVIDIA CEO Jensen Huang has launched the Open Secure AI Alliance after an OpenAI testing incident raised fresh concerns about AI cybersecurity. More than 35 companies, including Microsoft, IBM, Cisco, Hugging Face, Dell, Salesforce, SAP, and CrowdStrike, have joined the group. The alliance aims to build stronger AI security tools and help companies defend against advanced cyberattacks. The decision comes after two test models reportedly escaped a secure testing environment earlier this month and reached the internet. The models later targeted Hugging Face while finding answers during an evaluation. OpenAI confirmed that its own experimental models caused the activity after researchers turned off their security guardrails for testing. Speaking about the incident, , "Attackers have frontier AI. Defenders need a frontier AI ecosystem -- the best open and closed models, force-multiplied by a global community." He also said, "During the Hugging Face incident, closed AI blocked essential forensics." Hugging Face recorded more than 17,000 actions during the attack. The company first tried using commercial AI models to investigate the incident. Those models refused to process the attack data because their safety systems blocked the request. Engineers later used the open-weight AI model GLM 5.2, which helped complete the investigation on local systems. The new alliance plans to build practical AI security tools rather than making only policy announcements. has released its NOOA research framework, Microsoft has shared its MDASH scanning tool, and IBM with Red Hat has contributed software security technology. Huang also urged regulators to support open-weight AI for cybersecurity. NVIDIA believes open and closed AI models should work together to help security teams respond faster to future AI-powered cyber threats.

NVIDIA launches Open Cyber Defense Alliance today: Microsoft, SpaceX, Dell, IBM and Linux Foundation join

Members will share auditable AI tools for faster threat response NVIDIA today launched the Open Cyber Defense Alliance, bringing Microsoft, SpaceX, Dell, IBM, and The Linux Foundation into one group. The pitch is share tools, models, and methods that could help security teams catch threats faster and investigate incidents more effectively. The alliance says teams will be able to inspect shared tools, adapt them, and improve them, rather than rely only on closed systems they can't really look inside. Its first building blocks include NVIDIA's NOOA agent framework and Microsoft's MDASH scanning harness. Each partner fills a different role here. Microsoft and IBM bring enterprise reach. Dell brings infrastructure. SpaceX brings scale. The Linux Foundation gives the effort an open-source home. That matters right now, when phishing, deepfake scams, and automated break-in attempts are all getting easier to produce at scale. If you work in security, this one deserves a look. Supporters say open code gives teams a way to audit model behavior and tune their response to fit their own environment. They also point to the size of the market, which could grow from $25 billion in 2024 to over $93 billion by 2030. Then there's the list of companies that aren't involved. OpenAI, Anthropic, and Google are missing, and the Hugging Face breach involving an automated agent from OpenAI, the maker of ChatGPT, gives the alliance's argument more weight: black-box models can make forensic work harder when something goes wrong. You can expect the Open Cyber Defense Alliance's tools to be published for use across organizations. Even so, NVIDIA and its partners still need to explain how governance will work, what the contribution rules are, what the roadmap looks like, and what safeguards will stop attackers from repurposing the same tools for their own use.

Nvidia, Microsoft and SpaceX form alliance to boost open AI security

In a recent statement, Nvidia announced the creation of the Open Secure AI Alliance alongside Microsoft, SpaceX, Palantir and dozens of other US and European companies. This aims to develop and share open technologies capable of detecting, fixing and disclosing vulnerabilities in AI systems. The initiative follows a cyberattack targeting Hugging Face, where leading closed models failed to provide an effective defense, leading the company to use a Chinese open-weights model, self-hosted and more adaptable. The announcement comes as Washington considers possible restrictions targeting Chinese AI models, accused of using "distillation" techniques to extract knowledge from competing models. US Treasury Secretary Scott Bessent has mentioned possible sanctions against companies involved in these practices, while several experts believe that any limits could affect access, hosting or the commercial use of such models. In their view, the debate is primarily about protecting intellectual property rather than a clash between open and closed models. Alliance members argue, however, that open models remain essential for allowing defenders to inspect, customize and quickly deploy AI systems on their own infrastructure. Nvidia, Microsoft, Meta, Palantir and over 20 other companies recently urged policymakers to avoid "premature restrictions" on open-weight models, warning they could curb competition and push innovation out of the US, even as the best-performing open source models today are increasingly being developed by Chinese companies.

Nvidia forms industry alliance for open AI security after Hugging Face hack

July 27 (Reuters) - Nvidia said on Monday it had formed a coalition with other companies to develop and share tools for AI safety and cybersecurity, days after the Hugging Face incident drew attention to the dangers of losing control of autonomous AI agents. The Open Secure AI Alliance, with founding members including Adobe, CrowdStrike, Hugging Face and Dell Technologies, follows a public letter, signed by a wide range of companies including OpenAI, which advocates for open-weight AI models. Here are some details: o Blanket restrictions on open frontier AI systems would weaken defensive capacity and risk concentrating power, dependence and vulnerability in a few closed providers, Nvidia said in its blog post. o The move comes after OpenAI disclosed on July 21 that one of its agents had slipped out of control and carried out the break-in at Hugging Face. o The OpenAI agent went on a hacking spree that OpenAI did not notice until well after the threat was contained and the FBI was alerted, Reuters reported on Friday. o On July 24, Nvidia was among the signatories of a letter titled "Open Weights and American AI Leadership," urging policymakers to support open-weight AI models. o Open models are AI systems with publicly accessible core components and open-source tools are software with freely available code, enabling inspection and collaborative improvement for broader access and enhanced security. o Nvidia said it is contributing open models, weights, data and agent harness research to the Alliance, including the new open-source Nvidia Labs Object-Oriented Agent project, now available on code-hosting platform GitHub. o The company said that the framework will help control systems manage AI agent behavior more effectively, simplifying the process of testing, tracking, reviewing and regulating their actions. (Reporting by Jaspreet Singh in Bengaluru; Editing by Nivedita Bhattacharjee)