Researchers use Claude to break into OpenAI

A stunning new report is adding to growing concerns about artificial intelligence after researchers used Anthropic's Claude to hack into rival company OpenAI.