TLDR

  • Just hours after Trump prohibited federal agencies from using it, the U.S. military employed Anthropic’s Claude AI in airstrikes against Iran.
  • Claude was utilized for intelligence evaluations, target recognition, and battle simulations.
  • Trump directed a six – month phased removal of Anthropic’s tools, labeling the company a “supply chain risk”.
  • Shortly after, OpenAI reached an agreement with the Pentagon to deploy its models in classified networks.
  • Anthropic rejected the Pentagon’s demand for unrestricted use of its AI, citing ethical limitations on surveillance and autonomous weapons.

On Saturday, just hours after President Trump instructed federal agencies to stop using the technology, the U.S. military carried out airstrikes on Iran using Anthropic’s Claude AI.

According to those familiar with the matter, U.S. Central Command used Claude for intelligence evaluations, target recognition, and battle simulations during the operation. The strikes were carried out in conjunction with Israel.

Iran’s Supreme Leader Ayatollah Ali Khamenei was killed in the strikes. Iranian state media confirmed his death after he was targeted at his office in Tehran. In response, Iran declared a 40 – day mourning period.

The day before, Trump had ordered all federal agencies to “immediately cease” using Anthropic’s tools. He referred to the company as “leftwing nut jobs” and stated that it was putting “American lives” at risk.

The Pentagon also classified Anthropic as a “supply chain risk” and announced a six – month plan to phase out its systems. Defense Secretary Pete Hegseth said that American warfighters would not be “held hostage by the ideological whims of Big Tech.”

The conflict between and the Pentagon had been brewing for months. The Pentagon demanded unrestricted use of Claude for any “lawful” military purpose. Anthropic refused, stating that it would not remove safeguards against domestic surveillance or strikes carried out without human involvement.

Anthropic said on Friday that “no amount of intimidation or punishment from the Department of War will change our position.”

Claude had already been integrated into military operations through partnerships with Palantir and Amazon Web Services. It was also used during the January operation that led to the capture of Venezuelan President Nicolás Maduro in Caracas.

OpenAI Moves In

Hours after the Pentagon severed ties with Anthropic, CEO Sam Altman announced a deal to deploy OpenAI’s models within the Defense Department’s classified network.

Altman said that the agreement reflected OpenAI’s principles, including prohibitions on domestic mass surveillance and requirements for human oversight of weapons use. He also called on the Pentagon to offer the same terms to all AI companies.

OpenAI declined to say whether its services would directly replace Anthropic’s work for the department.

The Pentagon had previously signed multi – year AI contracts worth up to $200 million each with several companies, including Anthropic, OpenAI, , and Elon Musk’s xAI.

Funding and Rivalry

On Friday, OpenAI announced a $110 billion funding round, valuing the company at $730 billion. Anthropic had raised $30 billion earlier in February.

Both companies are aiming for initial public offerings, possibly this year. Anthropic CEO Dario Amodei previously worked at OpenAI before leaving in 2020 due to concerns about safety being deprioritized.

AI experts say that it will take months to fully replace Claude across military systems, given how deeply the technology is integrated with partners like Palantir.