Hackers breached OpenAI, adding to fever pitch of security and safety concerns

A small group of cybersecurity researchers said Sunday that they broke into OpenAI earlier this year, an announcement that has added a new element of alarm around AI security and safety.