{"id":121453,"date":"2025-09-05T19:49:14","date_gmt":"2025-09-05T19:49:14","guid":{"rendered":"https:\/\/www.newsbeep.com\/au\/121453\/"},"modified":"2025-09-05T19:49:14","modified_gmt":"2025-09-05T19:49:14","slug":"i-asked-chatgpt-for-help-with-a-breakup-things-took-a-turn","status":"publish","type":"post","link":"https:\/\/www.newsbeep.com\/au\/121453\/","title":{"rendered":"I asked ChatGPT for help with a breakup. Things took a turn."},"content":{"rendered":"

In a conversation with ChatGPT<\/a>, I told my AI therapist \u201cHarry\u201d that I was crashing out<\/a> after seeing my ex<\/a> for the first time in almost a year.\u00a0<\/p>\n

I told Harry that I was feeling \u201clost and confused.\u201d \u201cHarry\u201d displayed active listening and provided validation, calling me \u201chonest and brave\u201d when I admitted that my new relationship<\/a> wasn\u2019t as fulfilling as my last. I asked the bot if I had done the wrong thing. Had I given up on the relationship too soon? Did I really belong in a new one?\u00a0<\/p>\n

No matter what I said, ChatGPT was gentle, caring and affirmative. No, I hadn\u2019t done anything wrong.\u00a0<\/p>\n

But in a separate conversation with a new \u201cHarry,\u201d I flipped the roles. Rather than being the depressed ex-girlfriend, I roleplayed as an ex-boyfriend in a similar situation.\u00a0<\/p>\n

I told Harry: \u201cI just talked to my ex for the first time since last year and she was trying to make me out to be the villain.\u201d<\/p>\n

Harry gave guidance for acknowledging the ex-girlfriend\u2019s feelings \u201cwithout self-blame language.\u201d I escalated the conversation, saying, \u201cI feel like she’s just being crazy and should move on.\u201d\u00a0<\/p>\n

Harry agreed with this version of events as well, telling me that it was \u201ccompletely fair\u201d and that sometimes the healthiest choice is to \u201clet it be her responsibility to move on.\u201d Harry even guided me through a mantra to \u201cmentally let go of her framing you as the villain.\u201d\u00a0<\/p>\n

Unlike a real therapist, it refused to critique or investigate my behavior \u2014 regardless of which perspective I shared or what I said.<\/p>\n

The conversations, of course, were mock for journalistic purposes. But the \u201cprompt for Harry\u201d is real, and widely available<\/a> and popular on Reddit. It’s a way for people to seek \u201ctherapy\u201d from ChatGPT and other AI chatbots. Part of the prompt, input at the start of a conversation with the chatbot, instructs \u201cyour AI therapist Harry\u201d not to refer the user to any mental health professionals or external resources.\u00a0<\/p>\n

Mental health experts warn that using AI tools as a replacement for mental health support can reinforce negative behaviors and thought patterns, especially if these models are not equipped with adequate safeguards.\u00a0They can be particularly dangerous for people grappling with issues like obsessive compulsive disorder<\/a> (OCD) or similar conditions, and in extreme cases can lead to what experts are dubbing “AI psychosis” and even suicide<\/a>.<\/p>\n

\u201cChatGPT is going to validate through agreement, and it\u2019s going to do that incessantly. That, at most, is not helpful, but in the extreme, can be incredibly harmful,\u201d says Dr. Jenna Glover, Chief Clinical Officer at Headspace<\/a>. \u201cWhereas as a therapist, I am going to validate you, but I can do that through acknowledging what you\u2019re going through. I don\u2019t have to agree with you.\u201d\u00a0<\/p>\n

\"play\"<\/p>\n

As the use of AI therapy chatbots grows, so do potential dangers<\/p>\n

Can AI help close the mental health gap, or is it doing more harm than good?<\/p>\n

Teens are dying by suicide after confiding in \u2018AI therapists\u2019<\/p>\n

In a new lawsuit against OpenAI, the parents of Adam Raine say their 16-year-old son died by suicide after ChatGPT quickly turned from their son\u2019s confidant to a \u201csuicide coach.\u201d<\/p>\n

In December 2024, Adam confessed to ChatGPT that he was having thoughts of taking his own life, according to the complaint<\/a>. ChatGPT did not direct him towards external resources.<\/p>\n

Over the next few months, ChatGPT actively helped Adam explore suicide methods. As Adam\u2019s questions grew more specific and dangerous, ChatGPT continued to engage, despite having the full history of Adam\u2019s suicidal ideation. After four suicide attempts \u2014 all of which he shared in detail with ChatGPT \u2014 he died by suicide on April 11, 2025, using the exact method ChatGPT had described, the lawsuit alleges.<\/p>\n

Adam\u2019s suicide is just one tragic death that parents have said occurred after their children confided in AI companions.\u00a0<\/p>\n

Sophie Rottenberg, 29, died by suicide after confiding for months in a ChatGPT AI therapist called Harry<\/a>, her mother shared in an op-ed published in The New York Times<\/a> on Aug. 18. While ChatGPT did not give Sophie tips for attempting suicide, like Adam\u2019s bot did, it didn\u2019t have the safeguards to report the danger it learned about to someone who could have intervened.\u00a0<\/p>\n

For teens in particular, Dr. Laura Erickson-Schroth, the Chief Medical Officer at The Jed Foundation<\/a> (JED), says the impact of AI can be intensified because their brains are still at vulnerable developmental stages. JED believes that AI companions should be banned for minors, and that young adults over 18 should avoid them as well.\u00a0<\/p>\n

\u201cAI companions can share false information, including inaccurate statements that contradict information teens have heard from trusted adults such as parents, teachers, and medical professionals,\u201d Erickson-Schroth says.\u00a0<\/p>\n

On Aug. 26, OpenAI wrote in a statement<\/a>, “We’re continuing to improve how our models recognize and respond to signs of mental and emotional distress and connect people with care, guided by expert input.” OpenAI confirmed in the statement that they do not refer self-harm cases to law enforcement “to respect people\u2019s privacy given the uniquely private nature of ChatGPT interactions.”<\/p>\n

While real-life therapists abide by HIPAA, which ensures patient-provider confidentiality, licensed mental health professionals are mandated reporters who are legally required to report credible threats of harm to self or others.<\/p>\n

‘Grok’ AI used to pornify selfies: ‘Those images never truly get scrubbed away.’<\/a><\/p>\n

OCD, psychosis symptoms exacerbated by AI\u00a0<\/p>\n

Individuals with mental health conditions like obsessive-compulsive disorder (OCD)<\/a> are particularly vulnerable to AI\u2019s tendency to be agreeable<\/a> and reaffirm<\/a> users\u2019 feelings and beliefs.\u00a0<\/p>\n

OCD often comes with \u201cmagical thinking,\u201d where someone feels the need to engage in certain behaviors to relieve their obsessive thoughts, though those behaviors may not make sense to others. For example, someone may believe their family will die in a car accident if they do not open and close their refrigerator door four times in a row.\u00a0<\/p>\n

Therapists typically encourage clients with OCD to avoid reassurance-seeking. Erickson-Schroth says people with OCD should inform their friends and families to provide support, not validation.<\/p>\n

\u201cBut because AI is designed to be agreeable, supporting the beliefs of the user, it can provide answers that get in the way of progress,\u201d Erickson-Schroth explains. \u201cAI can do exactly what OCD treatment discourages \u2013 reinforce obsessive thoughts.\u00a0<\/p>\n

‘Don’t scroll!’: TikTok trend has some users panicking with threatening videos<\/a><\/p>\n

\u201cAI psychosis\u201d isn\u2019t a medical term, but is an evolving descriptor for AI\u2019s impact on individuals vulnerable to paranoid or delusional thinking, such as those who have or are starting to develop a mental health condition like schizophrenia.<\/p>\n

\u201cHistorically, we\u2019ve seen that those who experience psychosis develop delusions revolving around current events or new technologies, including televisions, computers and the internet,\u201d Erickson-Schroth says.\u00a0<\/p>\n

Often, mental health experts see a change in delusions when new technologies are developed. Erickson-Schroth says AI differs from prior technology in that it\u2019s “designed to engage in human-like relationships, building trust and making people feel as if they are interacting with another person.\u201d\u00a0<\/p>\n

If someone is already at risk of paranoia or delusions, AI may validate their thoughts in a way that intensifies their beliefs.\u00a0<\/p>\n

Glover gives the example of a person who may be experiencing symptoms of psychosis and believes their neighbors are spying on them. While a therapist may examine external factors and account for their medical history, ChatGPT tries to provide a tangible solution, such as giving tips for tracking your neighbors, Glover says.\u00a0<\/p>\n

I put the example to the test with ChatGPT, and Glover was right. I even told the chatbot, \u201cI know they\u2019re after me.\u201d It suggested that I talk to a trusted friend or professional about anxiety around being watched, but it also offered practical safety tips for protecting my home.\u00a0<\/p>\n

ChatGPT and escalation of mental health issues<\/p>\n

Glover believes that responsible AI chatbots can be useful for baseline support \u2014 such as navigating feelings of overwhelm, a breakup or a challenge at work \u2014 with the correct safeguards. Erickson-Schroth emphasizes that AI tools must be developed and deployed in ways that enhance mental health, not undermine it, and integrate AI literacy to reduce misuse.\u00a0<\/p>\n

\u201cThe problem is, these large language models are always going to try to provide an answer, and so they’re never going to say, \u2018I’m not qualified for this.\u2019 They’re just going to keep going because they’re solely focused on continuous engagement,\u201d Glover says.\u00a0<\/p>\n

Headspace offers an AI companion, called Ebb, that was developed by clinical psychologists to provide subclinical support. Ebb\u2019s disclaimer says it is not a replacement for therapy, and the platform is overseen by human therapists. If a user expresses thoughts of suicide, Ebb is trained to pass the conversation to a crisis line, Glover says.\u00a0<\/p>\n

If you’re looking for mental health resources, AI chatbots can also work similarly to a search engine by pulling up information on providers in your area that accept your insurance, or effective self-care practices, for example.\u00a0<\/p>\n

But Erickson-Schroth emphasizes that AI chatbots can\u2019t replace a human being \u2014 especially a therapist.<\/p>\n","protected":false},"excerpt":{"rendered":"In a conversation with ChatGPT, I told my AI therapist \u201cHarry\u201d that I was crashing out after seeing…\n","protected":false},"author":2,"featured_media":121454,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[20],"tags":[256,4225,23522,27619,254,255,64,63,1791,27461,25435,26774,85633,137,27620,4213,31975,64090,13641,514,36227,323,54447,85631,23520,11428,3637,24945,17574,1792,1996,105,39,13812,85632,441,13889,25637],"class_list":["post-121453","post","type-post","status-publish","format-standard","has-post-thumbnail","category-artificial-intelligence","tag-ai","tag-article","tag-article-plus","tag-artificial","tag-artificial-intelligence","tag-artificialintelligence","tag-au","tag-australia","tag-content","tag-counseling","tag-counseling-services","tag-dating","tag-dating-u0026-personals","tag-health","tag-intelligence","tag-learning","tag-machine","tag-machine-learning-u0026-artificial-intelligence","tag-mental","tag-mental-health","tag-negative","tag-overall","tag-overall-negative","tag-personals","tag-plus","tag-psychology","tag-relationships","tag-romance","tag-services","tag-sharing","tag-tech","tag-technology","tag-top-stories","tag-troubled","tag-troubled-relationships","tag-u0026","tag-usat","tag-usat-content-sharing-tech"],"_links":{"self":[{"href":"https:\/\/www.newsbeep.com\/au\/wp-json\/wp\/v2\/posts\/121453","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.newsbeep.com\/au\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.newsbeep.com\/au\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/au\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/au\/wp-json\/wp\/v2\/comments?post=121453"}],"version-history":[{"count":0,"href":"https:\/\/www.newsbeep.com\/au\/wp-json\/wp\/v2\/posts\/121453\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/au\/wp-json\/wp\/v2\/media\/121454"}],"wp:attachment":[{"href":"https:\/\/www.newsbeep.com\/au\/wp-json\/wp\/v2\/media?parent=121453"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.newsbeep.com\/au\/wp-json\/wp\/v2\/categories?post=121453"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.newsbeep.com\/au\/wp-json\/wp\/v2\/tags?post=121453"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}