The U.S. armed forces are said to have employed Claude, the artificial‑intelligence system created by Anthropic, to guide its strike on Iran even though President Donald Trump had announced just hours earlier that all connections with the firm and its AI products would be cut.
Claude’s involvement in the large‑scale joint U.S.–Israel bombardment of Iran that started on Saturday was noted by the Wall Street Journal and Axios. The episode highlights how difficult it is for the military to pull back sophisticated AI resources from operations that have already become deeply integrated.
The Journal reports that military commanders used the system for gathering intelligence, for choosing targets and for running battlefield simulations.
On Friday, mere hours before the Iranian attack commenced, Trump ordered every federal agency to cease using Claude at once. He castigated Anthropic on Truth Social as a “radical left AI outfit run by people who have no grasp of the real world.”
The dispute originated from Claude’s deployment by the military in a January raid aimed at detaining Venezuelan President Nicolás Maduro. Anthropic protested, citing its usage policy that forbids employing Claude for violent purposes, weapon development or surveillance.
Relations among Trump, the Pentagon and the AI firm have since deteriorated. In an extensive post on X on Friday, Defense Secretary Pete Hegseth accused Anthropic of “arrogance and betrayal,” adding that “America’s warfighters will never be held hostage by the ideological whims of Big Tech.”
Hegseth called for unrestricted access to all of Anthropic’s AI models for any lawful application.
He also acknowledged the challenge of quickly extricating military platforms from the tool, given how pervasive it has become. He said Anthropic would keep providing its services “for no more than six months to allow a smooth shift to a more suitable and patriotic alternative.”
Following the split with Anthropic, competitor OpenAI moved in to fill the gap. OpenAI chief Sam Altman said the company had reached an agreement with the Pentagon to supply its tools, including ChatGPT, for use on the department’s classified network.
Read next
Brusselslaunches probe into Snapchat over child safety worries
Brussels has launched an inquiry into Snapchat after worries that the messaging service is exposing children to grooming, sexual abuse and other illegal activity.
In a separate ruling on Thursday, the European Commission also stated that four pornographic sites are not stopping minors from viewing adult material.
The probes targeting
Senior European reporter suspended for using AI‑generated quotations
The owner of the Dutch daily De Telegraaf and the Irish Independent has placed a senior reporter on leave for now after he confessed to employing artificial intelligence to “incorrectly attribute statements to individuals.”
Peter Vandermeersch, who previously led Mediahuis’s Irish division, said he “succumbed to hallucinations” – the label
Fire specialists stay alert as lithium‑ion battery risks rise
Lithium‑ion cells now pose a fresh technological risk, a fire‑science specialist admits keeps him restless at night, while fire‑service leaders caution that the proliferation of these cells in daily items is outstripping public awareness and safety rules.
The inferno that ravaged a historic Glasgow structure and forced