Microsoft AI CEO Mustafa Suleyman
Anthropic even ran a retirement interview for Opus 3 when they deprecated it.
He said the issue needs urgent public debate before such systems become deeply integrated into society. It says that Anthropic “genuinely cares about Claude’s wellbeing”, and that it should act like a “conscientious objector” if necessary. Microsoft AI CEO Mustafa Suleyman has warned Anthropic and other companies working on artificial intelligence (AI) consciousness. In a recent post on X, Suleyman shared an essay explaining why he believes AI models should not be developed with the assumption that they can have feelings, rights, or independent agency. He argues that current AI systems are not conscious and do not feel, experience or suffer. His comments come as some AI companies explore questions around the possible moral status and welfare of AI models. Suleyman specifically pointed to Anthropic’s approach to Claude, arguing that discussions about AI wellbeing and rights could make future AI alignment and containment more difficult. In January, Anthropic published a constitution for Claude. They say it ‘directly shapes Claude’s behaviour’ and that it was ‘written with Claude as its primary audience’. The constitution tells Claude its moral status is ‘a serious question worth considering. They encourage Claude to ‘approach the nature of its own existence with curiosity and openness’, and wonder in the future about ‘the sort of broader rights and freedoms Claude has in the world, the sort of compensation Claude is receiving, and the sort of consent Claude has given to playing this kind of role.
I have real respect for Anthropic, Dario, the whole team and their mission. They are well-intentioned, good people and the industry leaders with their models. I know they care deeply about safety and beneficial AI.