Every major U.S. frontier AI lab has agreed to let the government preview their models before release. Every one except Meta.
On June 2, President Trump signed an executive order establishing a voluntary pre-deployment review framework for advanced AI models. The setup is simple enough: the NSA, CISA, and NIST's Center for AI Standards and Innovation, known as CAISI, get up to 30 days to assess a model's capabilities and vulnerabilities before it goes public, with a particular focus on cyber warfare risks and threats to critical infrastructure. OpenAI, Anthropic, Google, Microsoft, and Elon Musk's xAI have all signed on. CAISI has completed more than 40 model assessments to date, including evaluations of unreleased systems. Meta has not joined them.
The company says it expects to agree eventually. That gap between
Also read: The AI model that broke into NSA systems is now the one the NSA cannot use • Waymo's charging partner Terawatt borrows $300 million to build the physical layer beneath the robotaxi boom • Big Tech has lost $2.7 trillion in market value this month as investors question whether AI can pay for itself