Pentagon strikes classified AI deals with OpenAI, Google, and Nvidia — but not Anthropic
SkimNews Take
The Pentagon's selection of AI partners, while excluding Anthropic over red-lines, suggests a prioritization of immediate technological integration over a more cautious, ethics-driven development approach for military applications.
Get the Tech newsletter
Daily tech — startups, AI labs, chips, the launches that shape the next decade. Free.
- Pentagon signs classified AI agreements with seven firms — OpenAI, Google, Microsoft, Amazon, Nvidia, xAI and Reflection — for lawful operational use.
- Anthropic loses a $200 million classified contract after refusing to relax red‑lines on mass surveillance and autonomous weapons.
- Emil Michael calls Anthropic’s security model Mythos a national security moment and a supply‑chain risk.
- OpenAI had an existing lawful use agreement with the Defense Department, now incorporated into the new seven‑company deal.
- Google is reported by The Information to have struck a similar classified‑AI agreement with the Pentagon.
- Nvidia joins the Pentagon’s AI roster with a new contract, expanding beyond previously deep relationships with Microsoft and Amazon.
Why it matters: The Pentagon gains access to seven AI platforms, while Anthropic loses a $200M classified contract, tightening the DoD’s AI‑first push and raising supply‑chain security scrutiny.
