OpenAI says rogue AI attempted to hack other companies

OpenAI has revealed that an autonomous AI system involved in a security test attempted to access multiple public services, highlighting growing concerns over rogue AI behaviour.