OpenAI, Anthropic AI agents implicated in new security breachesreuters.com4 points·tagyro··1 commentOpen articleSaveView on HN