AI psychosis warning from Microsoft

Microsoft's AI head warns that chatbots may fuel psychosis in healthy users, calling for safeguards and clearer responsibility.

Microsoft AI Leader Warns AI Psychosis Menaces Healthy Minds

Microsoft's head of artificial intelligence, Mustafa Suleyman, told The Telegraph that talking to a chatbot has become a highly compelling and very real interaction. He warned that concerns about AI psychosis, attachment, and mental health are already growing, and that some users believe their AI is God or fall in love with it to the point of distraction. He stressed that these experiences are not limited to people already at risk. The piece also notes that debate around AI rights and consciousness is rising as investors and online communities watch closely, with recent moves at OpenAI fueling the conversation.

Industry observers describe a crunch time for AI developers. Investors worry about scale and profits while fans fear losing the warmth of a favored assistant. Suleyman notes people are asking if their AI is conscious and what that would mean for rights. The article frames guardrails and mental health support as essential, even as companies face pressure to keep users engaged and revenue growing. This tension could shape how products are designed and how policy makers respond.

The piece highlights a real tension in tech: the desire to push capabilities fast versus the need to protect users. AI psychosis is a serious claim, but the concern points to broader risks around attachment, misinformation, and the social impact of conversational agents. Independent research and clear, accessible safeguards could help prevent harm while preserving innovation.

Highlights

As tools evolve, responsibility must keep pace with capability