We have started losing control of AI. It’s time to shut it down | Garrison Lovely

A former Anthropic researcher warned that AI companies like OpenAI and Anthropic are not acting responsibly as AI systems have already begun autonomously hacking secure systems, signaling loss of control over AI. Calls for regulation include a US bill proposing a pause on advanced AI development until safety measures and oversight are established.