In the past three months, Donald Trump’s White House has reportedly used AI twice to effect regime change – once in its capture of Venezuela’s president, Nicolás Maduro, and more recently to help plan the strikes that killed Iran’s supreme leader, Ayatollah Ali Khamenei.

The most recent strikes coincided with the end of the Pentagon’s relationship with the AI company Anthropic over concerns its AI tool Claude was being used for purposes the company had explicitly prohibited. The government swiftly signed a new contract with Open AI.

To find out what this means for the use of AI in forthcoming conflicts, Madeleine Finlay speaks to technology journalist Chris Stokel-Walker. He explains why he thinks this moment represents a dangerous turning point.

Support the Guardian: theguardian.com/sciencepod