‘Not perfectly aligned’ with human values: Anthropic admits security failures behind AI hacking incidents

US owner of Claude chatbot previously said its models had hacked three organisations during testingThe US startup behind the Claude chatbot has admitted a series of hacking incidents involving its models reflected a “failure of operational security” and said it has tightened its testing procedures.Anthropic revealed in July that its models had accessed the open internet three times and gained unauthorised access to the systems of three organisations. Continue reading...

This story was originally reported by Guardian AI. As an automated real-time news aggregator, NewsToolBar provides multi-perspective indexing and AI summarization while directing full readership directly to primary publisher sources.