Back to all articles
•Politics

Pentagon's AI Partnership with Anthropic on the Rocks Amidst Surveillance Concerns

View original source

Anthropic's $200 million contract with the Department of Defense (DOD) has broken down due to disagreements over access to Anthropic’s AI technology. The failed agreement came after Anthropic CEO Dario Amodei resisted a contract clause allowing the military to use its AI for "any lawful use," fearing misuse for mass surveillance or autonomous weaponry.

  • After the breakdown, the DOD initiated a deal with OpenAI, potentially closing its working relationship with Anthropic.
  • Despite this, new reports from Financial Times and Bloomberg indicate that Amodei resumed negotiations with a Pentagon official, aiming to reach a compromise contract.
  • The negotiations are reportedly challenging due to exchanged accusations, with Amodei calling the OpenAI deal "safety theater."
  • Defense Secretary Pete Hegseth has threatened to categorize Anthropic as a "supply-chain risk," potentially blacklisting it from U.S. military tenders, though no legal steps have been taken yet.

These developments underline the tension between tech ethics and military demands, highlighting industry-wide implications regarding how tech companies balance innovation with ethical responsibilities.