NVIDIA and more than three dozen industry partners announced on July 27, 2026 the launch of the Open Secure AI Alliance, a coalition organized under the Linux Foundation to build and share open technologies for AI safety and security. The founding group spans cloud computing, cybersecurity, enterprise software, open source foundations, and AI research, and its stated mission is to give defenders open frontier tools they can inspect, adapt, and run on their own infrastructure.<\/p>\n
Open source software underpins cloud computing, financial services, manufacturing, telecommunications, government, and internet services, the alliance said in its announcement. Cybersecurity ranks among the top three beneficiaries of that open foundation, and the new group is designed to extend that approach to AI. The alliance builds on the Linux Foundation’s Akrites initiative and on prior work by the OpenSSF community, and it will focus on remediating and disclosing vulnerabilities using open technologies.<\/p>\n
The launch framing draws a contrast between defenses locked inside a small number of opaque systems and defenses built on open models, harnesses, and tools that any defender can study, adapt, and deploy. The alliance argues that open models and open harnesses are essential for cybersecurity because they widen access to defensive capabilities, increase transparency for defenders, allow data to remain under the operator’s control, and complement frontier closed models with customizable local controls.<\/p>\n
The inaugural partner list includes 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.<\/p>\n
NVIDIA is contributing open models, model weights, data, and new agent harness research, and has released the NVIDIA Labs Object-Oriented Agent (NOOA) project on GitHub. NOOA is a research framework for integrating harnesses with models so agent behavior can be tested, traced, audited, and governed. The alliance is also coordinating an open defense stack for agents that covers identity, isolation, safe model formats, multi-model scanning, and secure coding workflows.<\/p>\n
Specific contributions from other members include:<\/p>\n
Alliance organizers pointed to a recent security incident at Hugging Face as evidence for the approach. 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. The alliance’s argument is that defenders need the ability to inspect, adapt, and run advanced AI on their own infrastructure, and that closed tooling constrained that response at a critical moment.<\/p>\n
The alliance acknowledged that open models, like any powerful technology, can be misused, including through attempts to weaken safeguards or repurpose capabilities for cyber attacks. It argued those risks are not unique to open systems and must be managed wherever advanced AI is deployed, and that simply keeping weights closed does not prevent determined attackers from seeking or exploiting powerful AI. Its preferred response is to pair openness with strong safeguards, clear rules against malicious misuse, rigorous evaluation, and rapid remediation.<\/p>\n
The group is asking policymakers and regulators to treat open models, harnesses, and security tooling as defensive assets rather than liabilities. It warned that blanket restrictions on open frontier AI systems would weaken defensive capacity and risk concentrating power, dependence, and vulnerability in a few closed providers, and it called for public and private investment in shared open infrastructure for AI defense, including datasets, evaluation frameworks, attack simulators, and red-teaming tools.<\/p>\n
Alliance members expect governments, industry, and researchers to join the work, and they have framed the initiative as an ongoing effort rather than a one-time announcement. NVIDIA has separately opened registration for GTC Berlin, scheduled for October 20 to 22, where further technical work from the alliance is likely to surface.<\/p>\n
The Open Secure AI Alliance is a coalition announced on July 27, 2026 by NVIDIA and more than three dozen industry partners, organized under the Linux Foundation to develop and share open technologies, techniques, and tools for AI safety and security.<\/p>\n
Alliance members argue that open models and open harnesses let defenders inspect, adapt, and run advanced AI on their own infrastructure, which they say is essential for cyber defense and for avoiding dependence on a small number of closed providers.<\/p>\n
NVIDIA released the NOOA agent harness research framework on GitHub, Hugging Face donated the Safetensors format to the PyTorch Foundation, HPE is contributing to SPIFFE\/SPIRE for zero-trust identity, IBM and Red Hat are extending Lightwell for signed patches, Microsoft is contributing the MDASH multi-model agentic scanning harness, and SpacexAI has open sourced the Grok Build terminal-based coding agent with plans to release Grok model weights.<\/p>\n