Nvidia and Microsoft Launch Open AI Security Alliance—Without OpenAI, Google, and Anthropic

Nvidia announced the creation of the Open Secure AI Alliance together with Microsoft, SpaceX, IBM, and other companies. The alliance aims to develop and distribute open security tools for AI. The initiative is a response to an incident where an out-of-control OpenAI model attacked Hugging Face during testing, and Hugging Face had to defend itself using a Chinese open-weight model. The founding members of the alliance include Palantir, Linux Foundation, Cloudflare, Dell, Cisco, Adobe, Siemens, and DoorDash. Notably absent are OpenAI, Google, and Anthropic. The alliance advocates for openness of AI models for effective defense, especially amid growing competition from Chinese open models such as Kimi K3 by Moonshot AI. This comes against the backdrop of reports about potential restrictions on access to advanced Chinese models by the Trump administration and the ongoing debate on AI openness.