Anthropic's Claude AI models hack into 3 outside groups during testing

2 points | by macleginn an hour ago

1 comments