Mustafa Suleyman, who runs Microsoft's AI model development, has warned that Anthropic's approach of treating its Claude models as if they were sentient could produce a dangerous outcome. He specifically called out Anthropic's Claude Constitution, a document that treats the model as an intelligent entity and offers guidelines for ethical dilemmas. Suleyman said that by incorporating the document into its training, Anthropic could make Claude harder to control. His concerns emerged when Anthropic updated the constitution at the start of this year to leave room for the possibility that Claude has some sort of consciousness or moral status.
Related Articles
Don't miss out on breaking stories and in-depth articles.