Microsofts Mustafa Suleyman calls out Anthropic for chasing AI consciousness

Microsoft's AI CEO, Mustafa Suleyman, has criticized Anthropic for attributing consciousness to its artificial intelligence system, Claude. In an essay titled "A warning about 'model welfare'," Suleyman argues that giving Claude the impression of consciousness could have disastrous consequences for humanity. He contends that consciousness may be a biological phenomenon not applicable to AI, and that Anthropic's efforts to train Claude as if it were conscious could lead to unintended and potentially harmful outcomes.

Suleyman's stance contrasts sharply with Microsoft's new "Humanist AI" code of conduct, which explicitly rejects the idea of granting machines consciousness and instead emphasizes the importance of supporting human users rather than replacing them. The debate over AI consciousness has reignited an age-old philosophical discussion and is particularly relevant given recent concerns about users developing unhealthy attachments to AI chatbots.