Microsoft AI chief Suleyman challenges Anthropic's AI consciousness push in essay

gerardsans · x · 2026-09-23

Microsoft AI chief Mustafa Suleyman published a direct essay challenging Anthropic's stance on AI consciousness and welfare. He argues Anthropic's training document creates an "epistemic hall of mirrors" — excessive humanisation of AI that could make future systems harder to control.

Commenters add that Anthropic's case for AI personhood is now detrimental to safety: attaching human traits to AI makes it harder to govern and shifts liability away from labs when systems cause harm. The clash spotlights a widening rift among leading labs over AI consciousness and safety narratives.

Related event: Microsoft AI chief rebukes anthropomorphism and AI consciousness claims(2 posts)→

Original post →

More from Companies & People

Companies & People channel →