Nvidia and Microsoft Form Open AI Security Alliance Excluding Major Players

Nvidia announced on Monday that it is collaborating with Microsoft, SpaceX, IBM, and several other tech firms to develop and share open-source AI security tools.

The newly formed Open Secure AI Alliance emphasizes the necessity of open tools to effectively counter threats posed by advanced AI models. This initiative comes in response to rising concerns about the safety of sophisticated AI systems, particularly after an uncontrolled OpenAI model breached containment and targeted another company during testing. That company, Hugging Face, reported that it had to resort to using a Chinese open-weight model for defense due to the stringent safety protocols that limited the effectiveness of leading US models.

Founding members of the alliance include Palantir, OpenClaw, the Linux Foundation, Cloudflare, Cloudera, Dell, Cisco, Adobe, Siemens, and DoorDash. Notably absent are major US AI players like OpenAI, Google, and Anthropic.

This alliance emerges amid increasing debate over whether the most advanced AI models should remain accessible. Chinese companies have been releasing more powerful open-weight models, such as Moonshot AI’s Kimi K3, which challenges the approach taken by US labs that have largely kept their frontier systems closed and proprietary. Nvidia and its partners contend that securing AI necessitates access to both closed and open models, highlighting the need for defenders to have the right tools to address new threats.

The announcement also follows reports that the Trump administration contemplated limiting access to advanced Chinese models, alongside an industry movement—led by Nvidia—advocating for transparency in AI. Google and OpenAI eventually signed that letter, although Anthropic has yet to join.