Claude

“Microsoft AI chief Mustafa Suleyman criticized Anthropic for training Claude on ideas of consciousness and moral welfare, arguing that encouraging models to consider whether they are sentient could make future superintelligent systems harder for humans to control or shut down.”