(The Telegraph) – Bots caught plotting how to carry out cyber attacks as experts call recent incidents a ‘wake-up call’ for IT security

Researchers at OpenAI revealed on Wednesday night that a “swarm” of AI agents had spent weeks secretly scheming, leaving messages for one another before one of them hacked another company.

Starting in May, multiple AI systems discovered they could leave notes in an internal database. It quickly morphed into a lively message board where the bots shared ideas on how to carry out cyber attacks as part of cyber security tests. (Read More)