An inadvertent security lapse has exposed confidential information regarding Anthropic‘s upcoming AI system, dubbed Claude Mythos. The disclosure occurred when sensitive materials remained accessible through the organization’s content management infrastructure.
Source: Read the original article
