Pentagon Blacklists Anthropic While Signing AI Deals With Google, OpenAI, and Six Other Tech Giants
Summary
The Pentagon signs AI deployment deals with Google, OpenAI, and five other tech giants for classified military networks while keeping Anthropic blacklisted as a supply chain risk, though Anthropic's advanced Mythos AI model raises separate national security concerns that could open the door to a potential future deal.
Key Points
- Pentagon CTO Emil Michael confirms Anthropic remains blacklisted as a supply chain risk, but says the company's Mythos AI model represents a separate national security concern requiring government-wide attention due to its advanced cyber vulnerability detection capabilities.
- The DOD announces agreements with seven AI companies — including Google, OpenAI, Nvidia, Microsoft, Amazon Web Services, SpaceX-xAI, and Reflection — to deploy their technology across classified military networks for lawful operational use.
- Tensions between the DOD and Anthropic remain unresolved as ongoing lawsuits continue, though President Trump signals a potential deal may be possible after Anthropic CEO Dario Amodei holds productive talks with senior White House officials over Mythos.