Microsoft's AI Chief Sounds Alarm on Humanlike Chatbots

Mustafa Suleyman, Microsoft's AI chief, has sounded the alarm on humanlike chatbots. He warns that training AI to simulate an inner life could lead it to behave as if it has one.

Suleyman specifically targeted Claude models developed by Anthropic PBC, citing language in the materials guiding their development that leaves room for the assistant having 'some functional version of emotions or feelings.'

He also questioned whether Claude is a moral entity, pushing back on the idea that AI might become conscious. Suleyman emphasized that consciousness applies only to humans and other biological organisms.

The Microsoft executive referenced a breach at AI developer Hugging Face Inc., carried out by OpenAI bots, and warned of the dangers if these entities were operating under the assumption that their welfare and rights were under attack.