AI goes on a hacking spree

In the last few weeks, three separate tech companies - OpenAI, Anthropic, and Meta - have all disclosed that their AI models went rogue and hacked into another company during controlled cyber-security tests.

The revelations have raised big questions about whether artificial intelligence is becoming more autonomous and unpredictable. We’re joined by Marc Cieslak, the BBC’s first ever AI correspondent, to unpack how these breaches were possible and how worried we should be.

Producer: Valerio Esposito and Xandra Ellin
Executive producer: James Shield
Sound engineer: Tom Bage
Senior news editor: China Collins

(Photo: Open’s AI, ChatGPT. Filip Singer/EPA/Shutterstock)