Nvidia, Microsoft, SpaceX and IBM launched the Open Secure AI Alliance for shared open-source security tooling, notably without OpenAI, Google, or Anthropic.
Nvidia announced Monday it is joining Microsoft, SpaceX, IBM and other tech companies to build and share open-source AI security tools under a new Open Secure AI Alliance. The group argues that open tooling, not proprietary stacks, is required for effective AI security at scale.
The lineup is notable for who's missing: OpenAI, Google, and Anthropic — the three labs building the most capable frontier models — are not founding members, hinting at a split between infrastructure players and model labs on how AI security gets standardized.
Infrastructure vendors are moving to own the security layer before model labs do, which matters for enterprise buyers deciding whose tooling to standardize on. Expect competing security stacks to emerge along infra-vs-lab lines, forcing procurement teams to pick a side or build interoperability themselves.
The daily signal, curated. Get it in your inbox.
Subscribe on LinkedIn →