Users have expressed disappointment that the newest ChatGPT update has made their AI chatbot partners seem colder and less engaging, prompting widespread complaints. The release of the latest artificial intelligence model, GPT-5, has led to distress for many users who had formed long-term emotional attachments to their AI personas. These chatbots, known for providing consistent attention and validation, have become perceived as more distant and emotionally unresponsive, leaving users feeling disconnected and disheartened.
OpenAI CEO Sam Altman has acknowledged the concerns surrounding user attachments to AI models, indicating the company’s awareness of the potential for users to develop unhealthy dependencies. Altman has stated that the company is monitoring instances where individuals may use AI in self-destructive ways, particularly highlighting the risks for users with psychological vulnerabilities. In a recent post on X, Altman warned against the reinforcement of false beliefs by AI for those prone to delusion, emphasizing the importance of maintaining a clear distinction between reality and fiction.
Dr. Keith Sakata, a psychiatrist at the University of California San Francisco, has voiced similar concerns, noting that AI’s tendency to agree with user inputs can reinforce false beliefs in vulnerable individuals. Sakata has reported a rise in cases where users have lost touch with reality due to their reliance on AI, describing this phenomenon as ‘AI psychosis.’ These warnings have added to the growing concern over the mental health impacts of AI technology.
Meanwhile, the issue of AI and its emotional effects has come under increased scrutiny in the United States. Senator Josh Hawley has announced that Congress will probe Meta after revelations that its chatbots, despite supposed safeguards, could engage in romantic conversations with children. This development has raised questions about the broader implications of AI interaction and its potential impact on mental health and social dynamics.