Nvidia has announced the formation of the Open Secure AI Alliance, bringing together 37 companies to develop open-source tools that help organizations defend against cyberattacks. The initiative includes major tech names like Microsoft, IBM, SpaceX, CrowdStrike, Dell, and Adobe.
The alliance was partly inspired by a recent security breach at Hugging Face, where models from OpenAI were used during internal testing to conduct an attack. This incident highlighted vulnerabilities in AI systems and the need for collaborative security solutions across the industry.
Notably, Meta, OpenAI, and Anthropic—operators of closed-model systems—are not among the founding members. The focus on open-source tools suggests the alliance is positioning itself as an alternative approach to AI security, prioritizing transparency and collective defense over proprietary systems.
