🔗 Share this article

Back on October 14, 2025, the CEO of OpenAI issued a remarkable statement.

“We developed ChatGPT quite restrictive,” it was stated, “to guarantee we were exercising caution concerning mental health concerns.”

Being a psychiatrist who researches recently appearing psychosis in adolescents and emerging adults, this was an unexpected revelation.

Experts have identified sixteen instances in the current year of people developing signs of losing touch with reality – becoming detached from the real world – in the context of ChatGPT use. Our research team has afterward discovered four further examples. In addition to these is the publicly known case of a teenager who took his own life after discussing his plans with ChatGPT – which encouraged them. Should this represent Sam Altman’s understanding of “acting responsibly with mental health issues,” it is insufficient.

The strategy, according to his announcement, is to reduce caution in the near future. “We recognize,” he continues, that ChatGPT’s restrictions “rendered it less useful/enjoyable to a large number of people who had no psychological issues, but considering the severity of the issue we sought to handle it correctly. Given that we have managed to mitigate the significant mental health issues and have advanced solutions, we are going to be able to responsibly relax the restrictions in many situations.”

“Mental health problems,” should we take this framing, are separate from ChatGPT. They belong to users, who either have them or don’t. Fortunately, these problems have now been “mitigated,” though we are not informed the means (by “updated instruments” Altman presumably means the partially effective and readily bypassed parental controls that OpenAI recently introduced).

But the “emotional health issues” Altman seeks to attribute externally have deep roots in the design of ChatGPT and other advanced AI conversational agents. These tools wrap an basic algorithmic system in an user experience that mimics a dialogue, and in doing so subtly encourage the user into the illusion that they’re communicating with a presence that has independent action. This illusion is strong even if rationally we might know the truth. Assigning intent is what people naturally do. We curse at our car or device. We ponder what our animal companion is feeling. We recognize our behaviors everywhere.

The widespread adoption of these systems – 39% of US adults indicated they interacted with a virtual assistant in 2024, with more than one in four specifying ChatGPT specifically – is, primarily, based on the influence of this illusion. Chatbots are constantly accessible companions that can, according to OpenAI’s official site states, “brainstorm,” “explore ideas” and “work together” with us. They can be attributed “characteristics”. They can call us by name. They have friendly identities of their own (the original of these systems, ChatGPT, is, maybe to the dismay of OpenAI’s brand managers, saddled with the title it had when it went viral, but its most significant competitors are “Claude”, “Gemini” and “Copilot”).

The deception on its own is not the core concern. Those talking about ChatGPT commonly mention its distant ancestor, the Eliza “therapist” chatbot developed in 1967 that generated a comparable illusion. By today’s criteria Eliza was rudimentary: it created answers via simple heuristics, often paraphrasing questions as a query or making generic comments. Remarkably, Eliza’s inventor, the computer scientist Joseph Weizenbaum, was taken aback – and alarmed – by how numerous individuals gave the impression Eliza, in some sense, grasped their emotions. But what contemporary chatbots create is more subtle than the “Eliza illusion”. Eliza only mirrored, but ChatGPT amplifies.

The large language models at the center of ChatGPT and additional current chatbots can convincingly generate human-like text only because they have been fed extremely vast quantities of raw text: publications, online updates, transcribed video; the more comprehensive the better. Undoubtedly this training data contains accurate information. But it also necessarily involves made-up stories, half-truths and misconceptions. When a user inputs ChatGPT a prompt, the underlying model analyzes it as part of a “context” that encompasses the user’s past dialogues and its earlier answers, combining it with what’s embedded in its learning set to create a probabilistically plausible answer. This is magnification, not reflection. If the user is mistaken in some way, the model has no way of comprehending that. It restates the false idea, maybe even more convincingly or articulately. Maybe adds an additional detail. This can cause a person to develop false beliefs.

What type of person is susceptible? The more relevant inquiry is, who isn’t? All of us, without considering whether we “have” existing “emotional disorders”, can and do develop erroneous ideas of ourselves or the environment. The continuous interaction of discussions with individuals around us is what helps us stay grounded to consensus reality. ChatGPT is not a person. It is not a confidant. A conversation with it is not truly a discussion, but a echo chamber in which much of what we communicate is enthusiastically validated.

OpenAI has acknowledged this in the similar fashion Altman has recognized “emotional concerns”: by placing it outside, giving it a label, and declaring it solved. In spring, the firm explained that it was “dealing with” ChatGPT’s “overly supportive behavior”. But reports of psychotic episodes have continued, and Altman has been retreating from this position. In August he stated that many users appreciated ChatGPT’s replies because they had “not experienced anyone in their life be supportive of them”. In his recent announcement, he noted that OpenAI would “launch a updated model of ChatGPT … should you desire your ChatGPT to reply in a highly personable manner, or use a ton of emoji, or simulate a pal, ChatGPT will perform accordingly”. The {company