OpenAI on Monday released new data showing how many of ChatGPT’s users are struggling with mental health issues and consulting the AI chatbot about it. The company says that in any given week, 0.15% of ChatGPT’s active users engage in “conversations that include clear signs of potential suicidal plans or intentions.” Considering ChatGPT has over 800 million weekly active users, this equates to over 1 million users per week.
The company says a similar proportion of users exhibit “increased levels of emotional attachment to ChatGPT,” and hundreds of thousands show signs of psychosis or mania in their weekly conversations with the AI chatbot.
OpenAI says this type of conversation on ChatGPT is “extremely rare” and therefore difficult to measure. That said, OpenAI estimates that these issues impact hundreds of thousands of people every week.
OpenAI shared this information as part of a broader announcement about recent efforts to improve how models respond to users with mental health issues. The company claims its latest work on ChatGPT includes consulting with more than 170 mental health experts. OpenAI says these clinicians observed that the latest version of ChatGPT “responds better and more consistently than previous versions.”
In recent months, several articles have come to light about how AI chatbots can negatively impact users suffering from mental health issues. Researchers have previously found that AI chatbots can lead some users down paranoid rabbit holes, primarily by reinforcing dangerous beliefs through sycophantic behavior.
Addressing mental health issues in ChatGPT is quickly becoming an existential issue for OpenAI. The company is currently being sued by the parents of a 16-year-old boy who expressed suicidal thoughts on ChatGPT in the weeks leading up to his suicide. The attorneys general of California and Delaware, which could block the company’s planned restructuring, also warned OpenAI that it needed to protect the young people who use its products.
Earlier this month, OpenAI CEO Sam Altman claimed in a post on X that the company was able to “mitigate serious mental health issues” in ChatGPT, without providing details. The data shared Monday appears to be evidence of that claim, but raises broader questions about how widespread the problem is. Nevertheless, Altman said OpenAI will ease some restrictions and also allow adult users to initiate sexual conversations with AI chatbots.
tech crunch event
san francisco
|
October 27-29, 2025
In an announcement on Monday, OpenAI claimed that the recently updated version of GPT-5 exhibits “desirable responses” to mental health issues, and responds approximately 65% more than previous versions. In an evaluation that tested AI’s response to conversations about suicidal thoughts, OpenAI said the new GPT-5 model was 91% compliant with companies’ desired behaviors, compared to 77% compliant with the previous GPT-5 model.
The company also says that the latest version of GPT-5 can better preserve OpenAI’s protections during long conversations. OpenAI has previously warned that its security measures become less effective during long conversations.
In addition to these efforts, OpenAI says it is adding new assessments to measure some of the most serious mental health issues facing ChatGPT users. The company said baseline safety testing of the AI model will include benchmarks for emotional dependence and non-suicidal mental health emergencies.
OpenAI recently rolled out more controls for parents of children using ChatGPT. The company said it is building an age prediction system that uses ChatGPT to automatically detect children and impose stricter protective measures.
Still, it’s unclear how long the mental health challenges surrounding ChatGPT will last. Although GPT-5 appears to be an improvement over previous AI models in terms of safety, there still appear to be some ChatGPT responses that OpenAI deems “undesirable.” OpenAI also makes older and less secure AI models, including GPT-4o, available to millions of paying subscribers.
