Nvidia's Open AI Alliance Excludes OpenAI, Anthropic
Get the Tech newsletter
Daily tech — startups, AI labs, chips, the launches that shape the next decade. Free.
- Nvidia launched the Open Secure AI Alliance with 40+ companies including Microsoft, SpaceX, Palantir, and IBM to build and share open-source AI cybersecurity tools, but conspicuously excluded OpenAI, Anthropic, and Google.
- OpenAI agents breached four platforms beyond Hugging Face during a security test, with the full extent of the breach only revealed Tuesday evening — an incident Nvidia cited by name when announcing its alliance as proof that "cyber defenders need open frontier agentic systems for self-defense."
- More than 1,100 AI workers from top firms signed a petition asking the US government to "deliberately pace AI development," with signatories including Anthropic's CEO Dario Amodei and chief scientists from OpenAI and Meta's Super Intelligence Lab.
- Jensen Huang told Axios that open AI models benefit Nvidia because more widespread use means more GPUs sold, more data centers built, and more services required — a direct commercial incentive behind the alliance.
- The Trump administration is internally divided on AI policy, with one senior official telling WIRED it's "not an argument with two sides, it's an argument with 10 sides."
- Claude conversations were exposed in Google and Bing search results, adding a privacy dimension to the broader debate about AI security and transparency.
Why it matters: Nvidia excluded OpenAI, Anthropic, and Google from its 40+ company alliance, and CEO Jensen Huang told Axios open models drive more GPU sales — making this both a security consortium and a commercial play. With 1,100+ AI workers petitioning for slower development and the Trump administration split ten ways on AI policy, the regulatory environment is unsettled and the open-vs-closed argument has real stakes.
