OpenAI warns of ‘potentially catastrophic’ risks from superintelligent AI, outlines global safety measures

OpenAI also called for coordinated global oversight as the AI industry gets closer to self-improvement milestone.