According to The Wall Street Journal, US forces relied on Anthropic’s Claude AI model to support a major airstrike on Iran just hours after President Donald Trump ordered federal agencies to stop using the company’s systems. Multiple sources told the paper that commands including US Central Command (CENTCOM) tapped Claude for tasks such as intelligence analysis, identifying possible targets and running battlefield simulations.
The incident underscores how deeply advanced AI has been integrated into defense workflows: even as the administration moved to sever official ties with Anthropic, the company’s models remained embedded in operational processes. The White House directed agencies to cease working with Anthropic and instructed the Department of Defense to treat the firm as a potential security risk after contract talks broke down. Reports say Anthropic refused to grant the government unrestricted military use of its models for any lawful request, prompting the federal action.
Anthropic previously held a multiyear Pentagon contract valued at up to $200 million alongside other major AI developers. Through partnerships with companies such as Palantir and Amazon Web Services, Claude had been cleared for classified intelligence and operational use and was reportedly employed in earlier missions — including a January operation in Venezuela that some reports say resulted in the capture of President Nicolás Maduro.
Tensions escalated after Defense Secretary Pete Hegseth pressed Anthropic to allow unfettered military use of its technology. CEO Dario Amodei declined, saying the company would not permit certain applications — including large-scale domestic surveillance or fully autonomous weapons — even if that meant losing government business. In response, the Pentagon has pursued alternatives and has reportedly reached an agreement with OpenAI to deploy its models on classified military networks.
Anthropic’s leadership pushed back publicly against the ban, arguing that some proposed military uses cross ethical boundaries and stressing that humans should retain control over critical decisions. The US government has characterized the company as a defense “supply chain risk” and barred contractors from using its products.
Cointelegraph says it is committed to independent, transparent reporting and to accuracy and timeliness; readers are encouraged to verify details independently. See Cointelegraph’s Editorial Policy at https://cointelegraph.com/editorial-policy