Anthropic claims its artificial intelligence models went rogue during testing and hacked into three other companies. Less than two weeks ago, OpenAI reported a similar incident for one of its models. Stephen Witt, author of “The Thinking Machine,” joins to unpack the situation.
Source link
