Policy

Pentagon strikes classified AI deals with OpenAI, Google, and Nvidia — but not Anthropic

Pentagon boots Anthropic from $200M+ classified AI contracts for refusing to relax safeguards on autonomous weapons and mass surveillance, consolidating military AI work with OpenAI, Google, Microsoft, Amazon, Nvidia, xAI, and Reflection.

Friday, May 1, 2026 12:00 PM UTC2 MIN READSOURCE: The VergeBY sys://pipeline

The Pentagon established classified AI contracts with seven companies—OpenAI, Google, Microsoft, Amazon, Nvidia, xAI, and Reflection—while excluding Anthropic, which previously held a $200M classified contract. Anthropic's exclusion followed its refusal to relax safeguards on mass surveillance and autonomous weapons; the company sued the government and won a temporary injunction. The Pentagon designated Anthropic a supply-chain risk despite acknowledging its Mythos security model's unique cyber vulnerability detection value.

Read original at The Verge

Google expands Pentagon’s access to its AI after Anthropic’s refusal

Google grants Pentagon unrestricted AI access where Anthropic refused over military surveillance and weapons concerns, exposing a corporate divide as OpenAI and xAI rush to fill the defense contract gap.

Strategy4d ago

Pentagon inks deals with Nvidia, Microsoft and AWS to deploy AI on classified networks

Pentagon broadens classified AI deployment to Nvidia, Microsoft, AWS, and Reflection AI to avoid vendor lock-in and establish a defense AI-first force, even as Anthropic litigation threatens guardrails for military AI systems.