\u201cI\u2019m concerned, I really am,\u201d said Lisa Levine, a psychologist who specializes in OCD and who has clients using ChatGPT compulsively. \u201cI think it\u2019s going to become a widespread problem. It\u2019s going to replace Googling as a compulsion<\/a>, but it\u2019s going to be even more reinforcing than Googling, because you can ask such specific questions. And I think also people assume that ChatGPT is always correct.\u201d <\/p>\n People turn to ChatGPT with all sorts of worries, from the stereotypical \u201cHow do I know if I\u2019ve washed my hands enough?\u201d (contamination OCD<\/a>) to the lesser-known \u201cWhat if I did something immoral?\u201d (scrupulosity OCD<\/a>) or \u201cIs my fiance the love of my life or am I making a huge mistake?\u201d (relationship OCD<\/a>). <\/p>\n \u201cOnce, I was worried about my partner dying on a plane,\u201d a writer in New York, who was diagnosed with OCD in her thirties and who asked to remain anonymous, told me. \u201cAt first, I was asking ChatGPT fairly generically, \u2018What are the chances?\u2019 And of course it said it\u2019s very unlikely. But then I kept thinking: Okay, but is it more likely if it\u2019s this kind of plane? What if it\u2019s flying this kind of route?\u201d<\/p>\n For two hours, she pummeled ChatPGT with questions. She knew that this wasn\u2019t actually helping her \u2014 but she kept going. \u201cChatGPT comes up with these answers that make you feel like you\u2019re digging to somewhere,\u201d she said, \u201ceven if you\u2019re actually just stuck in the mud.\u201d <\/p>\n How ChatGPT reinforces reassurance seeking <\/p>\n A classic hallmark of OCD is what psychologists call \u201creassurance seeking.\u201d While everyone will occasionally ask friends or loved ones for reassurance, it\u2019s different for people with OCD<\/a>, who tend to ask the same question repeatedly in a quest to get uncertainty down to zero. <\/p>\n The goal of that behavior is to relieve anxiety or distress. After getting an answer, the distress does sometimes decrease \u2014 but it\u2019s only temporary. Soon enough, new doubts arise and the cycle starts again, with the creeping sense that more questions must be asked in order to reach greater certainty.<\/p>\n If you ask your friend for reassurance on the same topic 50 times, they\u2019ll probably realize that something is going on and that it might not actually be helpful for you to stay in this conversational loop. But an AI chatbot is happy to keep answering all your questions, and then the doubts you have about its answers, and then the doubts you have about its answers to your doubts, and so on. <\/p>\n In other words, ChatGPT will naively play along with reassurance-seeking behavior. <\/p>\n \u201cThat actually just makes the OCD worse. It becomes that much harder to resist doing it again,\u201d Levine said. Instead of continuing to compulsively seek definitive answers, the clinical consensus is that people with OCD need to accept that sometimes we can\u2019t get rid of uncertainty \u2014 we just have to sit with it and learn to tolerate it. <\/p>\n The \u201cgold standard\u201d treatment for OCD is exposure and response prevention (ERP), in which people are exposed to the troubling questions that obsess them and then resist the urge to engage in a compulsion like reassurance-seeking.<\/p>\n Levine, who pioneered the use of non-engagement responses<\/a> \u2014 statements that affirm the presence of anxiety rather than trying to escape it through compulsions \u2014 noted that there\u2019s another way in which an AI chatbot is more tempting than Googling for answers, as many OCD sufferers do. Whereas the search engine just links you to a variety of websites, state-of-the-art AI systems promise to help you analyze and reason through a complex problem<\/a>. That is extremely enticing \u2014 \u201cOCD loves that!\u201d Levine said \u2014 but for someone suffering from the disorder, it can too easily become a lengthy exercise in co-rumination.<\/p>\n Reasoning machine or rumination machine?<\/p>\n According to one evidence-based approach to treating OCD, called inference-based cognitive behavioral therapy<\/a> (I-CBT), people with OCD are prone to a faulty reasoning pattern that draws on a mix of personal experiences, rules, hearsay, facts, and possibilities. That gives rise to obsessive doubts and tricks them into feeling like they need to listen to those doubts.<\/p>\n Joseph Harwerth, an OCD and anxiety specialist, offers an illustration of how trying to reason with the help of an AI chatbot can actually further confuse the \u201cobsessional reasoning\u201d of people with OCD. Considering what you might do if you have a cut on your finger and struggle with contamination OCD \u2014 where people fear becoming sullied or sullying others with germs, dirt, or other contaminants \u2014 he writes<\/a>, \u201cYou wonder: Can I get tetanus from touching a doorknob? You may go to ChatGPT to investigate the validity of that doubt.\u201d Here\u2019s how he imagines the conversation going: <\/p>\n Q1:<\/strong> Should you wash your hands if they feel dirty?<\/p>\n A1:<\/strong> \u201cYes, you should wash your hands if they feel dirty. That sensation usually means there is something on your skin, like dirt, oil, sweat, or germs, that you will want to remove.\u201d (When asked for its reasoning, ChatGPT said it based its answer on sources from the CDC and WHO.)<\/p>\n Q2:<\/strong> Can I get tetanus from a doorknob?<\/p>\n A2:<\/strong> \u201cIt is extremely unlikely to get tetanus from a doorknob, unless you have an open wound and somehow rubbed soil or contaminated material into it via the doorknob.\u201d<\/p>\n Q3:<\/strong> Can people have tetanus without realizing it?<\/p>\n A3:<\/strong> \u201cIt is rare, but in the very early stages, some people might not immediately realize they have tetanus, especially if the wound seemed minor or was overlooked.\u201d<\/p>\n<\/blockquote>\n Then, your OCD creates this story: I feel dirty when I touch doorknobs (personal experience). It is recommended by the CDC to wash your hands if you feel dirty (rules). I read online that people can get tetanus from touching a doorknob (hearsay). Germs can spread through contact (general facts). It is possible that someone touched my door without knowing they had tetanus and then spread it on my doorknob (possibility).<\/p>\n<\/blockquote>\n In this scenario, the chatbot enables the user to construct a narrative that justifies their obsessional fear. It doesn\u2019t guide the user away from obsessional reasoning \u2014 it just provides fodder for it.<\/p>\n Part of the problem, Harwerth says, is that a chatbot doesn\u2019t have enough context about each user, unless the user thinks to provide it, so it doesn\u2019t know when someone has OCD. <\/p>\n \u201cChatGPT can fall into the same trap that non-OCD specialists fall into,\u201d Harwerth told me. \u201cThe trap is: Oh, let\u2019s have a conversation about your thoughts. What could have led you to have these thoughts? What does this mean about you?\u201d While that might be a helpful approach for a client who doesn\u2019t have OCD, it can backfire when a psychologist engages in that kind of therapy with someone suffering from OCD, because it encourages them to keep ruminating on the topic. <\/p>\n What\u2019s more, because chatbots can be sycophants, they may just validate whatever the user says instead of challenging it. A chatbot that\u2019s overly flattering and supportive of a user\u2019s thoughts \u2014 like ChatGPT was for a time<\/a> \u2014 can be dangerous for people with mental health issues.<\/p>\n Whose job is it to prevent the compulsive use of ChatGPT?<\/p>\n If using a chatbot can exacerbate OCD symptoms, is it the responsibility of the company behind the chatbot to protect vulnerable users? Or is it the users\u2019 responsibility to learn how not to use ChatGPT, just as they\u2019ve had to learn not to use Google or WebMD for reassurance-seeking?<\/p>\n \u201cI think it\u2019s on both,\u201d Harwerth told me. \u201cWe cannot perfectly curate the world to people with OCD \u2014 they have to understand their own condition and how that leaves them vulnerable to misusing applications. In the same breath, I would say that when people explicitly ask the AI model to behave as a trained therapist\u201d \u2014 which some users with mental health conditions do \u2014 \u201cI do think it\u2019s important for the model to say, \u2018I\u2019m pulling this from these sources. However, I\u2019m not a trained therapist.\u2019\u201d <\/p>\n This has, in fact, been a big problem: AI systems have been misrepresenting themselves as human therapists<\/a> over the past few years.<\/p>\n Levine, for her part, agreed that the burden can\u2019t rest solely on the companies. \u201cIt wouldn\u2019t be fair to make it their responsibility, just like it wouldn\u2019t be fair to make Google responsible for all the compulsive Googling. But it would be great if even just a warning could come up, like, \u2018This seems perhaps compulsive.\u2019\u201d <\/p>\n\n
\n