Anthropic's Pentagon Deal May Revive Over AI Edge
Get the Tech newsletter
Daily tech — startups, AI labs, chips, the launches that shape the next decade. Free.
- Anthropic is suing the Trump administration after Defense Secretary Pete Hegseth declared it a "supply chain risk to national security" for refusing to allow Claude to power fully autonomous warfare or mass surveillance of Americans.
- Pentagon insiders say Anthropic's AI is "vastly better for warfare" than ChatGPT, Gemini, or Grok, and credit it with giving the U.S. a 6-12 month lead over China in defense AI, with Claude also being used extensively in the Iran war.
- Anthropic stands to lose "tens of billions of dollars" in direct and indirect government contracts if the blacklisting holds, while Hegseth's bottom line remains "all lawful uses" — a negotiator told Axios there's "nothing to talk about" otherwise.
- Emil Michael, the Pentagon official leading negotiations, wrote in a March 4 email — five days after the blacklisting — that the sides were "very close" on deal language, though he publicly called CEO Dario Amodei "a liar" with "a God-complex."
- Brad Gerstner of Altimeter Capital — an Anthropic investor, Amodei adviser, and architect of Trump Accounts — could serve as a middleman, with Anthropic potentially donating to Trump Accounts or compromising on Pentagon-use language to revive the deal.
Why it matters: Anthropic faces tens of billions in lost contracts from the blacklisting, and Pentagon insiders credit Claude with giving the U.S. a 6-12 month AI defense lead over China — putting financial and strategic pressure on both sides to revive talks despite Hegseth's 'all lawful uses' red line.
