An escalating row between the Pentagon and AI firm Anthropic threatens to derail a $200 million contract, with the company refusing to ease safeguards on its Claude model that could prevent deployment in mass surveillance or autonomous weapons.
Claude, the first such system approved for the U.S. military’s classified Replicator network according to Reuters reporting from 24 February 2026, has sparked this clash over ethical boundaries in defence technology.
Defence Secretary Pete Hegseth confronted Anthropic chief executive Dario Amodei on Tuesday, insisting the AI must serve “all lawful purposes” or risk contract cancellation and designation as a supply chain threat—a status usually reserved for foreign adversaries, as Pentagon officials outlined to CNN.
Anthropic rejected the proposed changes outright, viewing them as a sham compromise riddled with legal loopholes that could bypass protections whenever convenient.
Amodei expanded on the firm’s position in Thursday’s lengthy blog post, emphasising his belief in AI’s role for national defence. “I believe deeply in the existential importance of using AI to defend the United States and other democracies, and to defeat our autocratic adversaries.”
While acknowledging that military choices belong to the Pentagon rather than private firms, he warned that specific applications like widespread monitoring and self-governing arms exceed what current technology can handle reliably, potentially undermining democratic values instead of upholding them.
These restrictions, he noted, have not impeded the armed forces’ use of Anthropic models to date, and official pressure will not sway them. “We cannot in good conscience accede to their request.”
Pentagon Undersecretary for Research and Engineering Emil Michael, who joined the talks, responded sharply on X. “It’s a shame that @DarioAmodei is a liar and has a God-complex. He wants nothing more than to try to personally control the US Military and is ok putting our nation’s safety at risk. The @DeptofWar will ALWAYS adhere to the law but not bend to whims of any one for-profit tech company.”
Anthropic employees quickly voiced support online, with technical alignment specialist Trenton Bricken posting, “Time and time again over my three year tenure at Anthropic I’ve seen us stand to our values in ways that are often invisible from the outside. This is a clear instance where it is visible.”
Data science manager Gian Segato echoed this, writing, “[H]istory is unfolding in front of us it’s now obvious and evident to everyone with eyes to see why anthropic founding was a crucial fork in the timeline, and how catastrophic the counterfactual would’ve been otherwise.”