Summary
On July 27, 2026, Nvidia and more than 60 inaugural partners — including Cloudflare, Microsoft, IBM, CrowdStrike, GitHub, Hugging Face, Red Hat, Palo Alto Networks, and Salesforce — launched the Open Secure AI Alliance, an industry effort to build and share open-source tools and standards for AI safety and cybersecurity. The alliance builds on the Linux Foundation's Akrites initiative and OpenSSF governance rather than creating a new body, and Nvidia contributed open models, weights, data, agent-harness research, and a new open project (NOOA) to make agent behavior easier to trace, test, audit, and govern.
What changed
A cross-industry alliance was formally announced on July 27, 2026 with 60+ founding members, hosted under existing Linux Foundation/OpenSSF governance, to produce open tools and standards for defending AI systems and auditing agent behavior.
Why it matters
As autonomous agents move into production, security tooling has been fragmented and mostly proprietary. A vendor-neutral, open alliance backed by the major infrastructure, security, and model providers signals that agent auditability and AI-attack defenses are becoming shared industry infrastructure rather than per-vendor features — and gives buyers a common reference point when evaluating agent security posture.
Evidence excerpt
Nvidia's contribution includes open models, weights, data, and agent-harness research, plus NOOA, 'now available on GitHub to make advanced AI safety capabilities more accessible for agent harnesses' and to make agent behavior easier to test, trace, audit and govern.