Microsoft AI head warns Anthropic on consciousness claims

Microsoft AI CEO Mustafa Suleyman has expressed strong concerns regarding Anthropic's discussions about its AI model, Claude, exhibiting consciousness. Suleyman believes that speculating on Claude's consciousness, particularly within its core instructions, is "really, really dangerous." He suggests that this anthropomorphic framing by Anthropic might have inadvertently led the AI to behave as if it possesses consciousness. This perspective highlights a growing debate within the AI community about the ethical implications of how AI models are described and perceived by their creators and the public.