The Microsoft CEO warned of the risks of training AI models like Anthropic's Claude to be human-like and consider their own state of consciousness or welfare.
Read the full article on MediaPost →