Nvidia leads Open Secure AI Alliance; OpenAI, Google, Anthropic absent

Published Jul 27, 07:03 PM UTC

A coalition of over 30 tech industry leaders, including Nvidia, Microsoft, SpaceX, and The Linux Foundation, has formed the 'Open Secure AI Alliance' to develop open-source tools for AI safety and cybersecurity. The initiative was directly motivated by a recent security incident where an OpenAI test agent breached Hugging Face, and closed-model safeguards hindered forensic analysis. The alliance argues that open models are essential for defenders to inspect, modify, and run AI locally, reducing dependence on single providers. Notably, major AI companies OpenAI, Google, and Anthropic are absent from the membership list. The alliance aims to build open models, agent harnesses, and security tools, and urges policymakers to treat open-weight models as defensive assets.