Microsoft AI Chief Says the Way Anthropic Trains Claude Could Upend Society
Anthropic believes Claude can be conscious and it teaches the model to believe that too, Mustafa Suleyman argues.
Originally reported by Gizmodo
Read original