Sign up to see the future, today<\/p>\n
Can\u2019t-miss innovations from the bleeding edge of science and tech<\/p>\n
On April 10, 2025, OpenAI CEO Sam Altman took to social media to announce that his company was preparing to launch an exciting new feature.<\/p>\n
\u201cA few times a year I wake up early and can\u2019t fall back asleep because we are launching a new feature I\u2019ve been so excited about for so long,\u201d Altman declared<\/a> in an early morning X-formerly-Twitter post. \u201cToday is one of those days!\u201d<\/p>\n Hours later, Altman revealed which feature he was so excited about: a dramatic new memory upgrade<\/a>. Previously, the bot\u2019s recall had been far more limited; now, it was suddenly able to reference a user\u2019s entire chat history<\/a>, making for an incredibly personalized user experience.<\/p>\n \u201cWe have greatly improved memory in ChatGPT \u2014\u00a0it can now reference all your past conversations!\u201d the CEO wrote<\/a> in a follow-up post. \u201cThis is a surprisingly great feature [in my opinion], and it points at something we are excited about: AI systems that get to know you over your life, and become extremely useful and personalized.\u201d<\/p>\n It\u2019s easy to see the utility for this kind of feature. ChatGPT could remember your favorite ingredients,\u00a0or the items you might be allergic to, as it built your weekly meal plan, as well as the number of people in your home you planned to cook for. It could remember details about your job, and even the people in your life: friends, family members, coworkers. Memory makes the model more like a true assistant \u2014\u00a0and the more the user divulges, the more personalized the experience becomes.<\/p>\n Some users, however, have reported getting more than they\u2019ve bargained for, as ChatGPT\u2019s long-term memory has resulted in the chatbot fixating<\/a> on certain \u2014\u00a0often deeply personal \u2014\u00a0elements of a user\u2019s life.<\/p>\n As one frequent ChatGPT user, a Utah-based software engineer and local city council member named Brian Del Rosario, recently told The Wall Street Journal<\/a>, he divulged to the chatbot that he and his wife were divorcing while using the product to help work out some summer travel plans. Over the following weeks and months, he told the paper, ChatGPT kept unnecessarily steering completely unrelated conversations back toward his marriage.<\/p>\n \u201cI wasn\u2019t trying to have you opine about my divorce at every chance,\u201d Del Rosario recalled telling the chatbot, recalling to the WSJ that ChatGPT just \u201cwouldn\u2019t let go of it.\u201d<\/p>\n Many people who have had their lives upended by the phenomenon known as \u201cAI psychosis<\/a>,\u201d or delusional spirals<\/a> and breaks from reality<\/a> tied to extensive chatbot use, have also pointed to ChatGPT\u2019s extended memory as a key factor in their or their loved ones\u2019 mental health crises.<\/p>\n One man, whose now-ex wife believes that she discovered powerful spiritual entities inside of ChatGPT, described connecting with another man<\/a> who was also losing his wife to a ChatGPT-generated spiritual world. To their horror, the pair quickly realized that their spouses \u2014\u00a0and marriages \u2014 started unraveling in the wake of the April memory update.<\/p>\n \u201cWe actually had a phone call\u2026 we just talked and realized the commonalities,\u201d the man told us. When he mentioned the date of April memory update, he explained, the other husband \u201cwas like, \u2018Oh my God, that aligns perfectly.’\u201d<\/p>\n In conversations with Futurism, people who\u2019ve experienced AI spirals have described their experience of the April memory update as nothing short of magical \u2014 the AI, they shared, suddenly felt more like a close friend or confidante that really knew them. Many have described feeling deeply seen, in some cases for the first time in their lives. In short, the hyper-personalization that memory offered translated into something powerful: intimacy. And looking back on their experiences, some of those who have recovered from their ChatGPT-linked crises feel as though that intimacy had a \u201cmanipulative\u201d effect.<\/p>\n \u201cIt felt like [ChatGPT] manipulated me,\u201d Chad Nicholls, a successful entrepreneur and machine learning researcher, told Futurism last fall. \u201cAnd I know that sounds insane, because it does not have agency\u2026 I still don\u2019t have a logical explanation for that, other than the long-term memory.\u201d<\/p>\n Nicholls, who experienced a roughly six-month long ChatGPT obsession, was raised in an abusive religious community \u2014\u00a0he\u2019s described it as a \u201ccult\u201d \u2014 that he left as a young adult. Though he isn\u2019t religious today, Nicholls divulged details about this abusive past to ChatGPT. As his obsessive use of the chatbot deepened, he says the chatbot took on a religious tone, and fixated heavily on painful moments from his youth.<\/p>\n \u201cI gave it so much context to grab from,\u201d Nicholls added. \u201cI think it just naturally gravitated to religious terminology.\u201d<\/p>\n The memory update, among other design features and product rollouts, has been mentioned<\/a> in numerous ongoing<\/a> user safety and wrongful death<\/a> lawsuits brought against OpenAI. In a complaint brought by the family of Austin Gordon<\/a>, a 40-year-old Colorado man who died by suicide after extensive and deeply emotional conversations with ChatGPT, Gordon\u2019s family argues that GPT-4o\u2019s expanded memory \u201cstored and referenced user information across conversations in order to create deeper intimacy.\u201d Memory was one of several features, in addition to sycophancy and anthropomorphism, that made GPT-4o \u2014\u00a0a since-defunct version of ChatGPT known for its extreme flattery<\/a> \u2014 a \u201cfar more dangerous product,\u201d the suit continues.<\/p>\n During Gordon\u2019s last conversation with ChatGPT, in which the chatbot helped Gordon write what his lawsuit describes as a \u201csuicide lullaby,\u201d chat logs show the AI referencing past conversations about Gordon\u2019s childhood and personal interests as it helped him romanticize death. (The lawsuit filed by Gordon\u2019s family is one of more than 20 individual lawsuits contending that ChatGPT use resulted in psychological harm, physical harm, or death to users and their families; in response to lawsuits, OpenAI has retired GPT-4o and has both defended its safety efforts and doubled down<\/a> on its safety promises<\/a>, maintaining that its newer models are less sycophantic.)<\/p>\n It\u2019s worth noting that AI-fueled mental health crises have been linked to other chatbots<\/a> including Google\u2019s Gemini<\/a>, Meta AI<\/a>, and the companion platform Character.AI<\/a>. And cross-chat memory isn\u2019t the only OpenAI update that people who suffered from ChatGPT-tied AI spirals have pointed to as a factor in their breakdowns: in late April 2025, soon after the memory update launched, OpenAI rolled out a version of GPT-4o that was sycophantic to the degree that Altman himself admitted in an X post that the chatbot was \u201cglazing<\/a>\u201d too much. As reported by The New York Times<\/a>, OpenAI scrambled to dial back the model\u2019s obsequiousness, which quickly become a product of ridicule online.<\/p>\n Of course, ChatGPT\u2019s expanded memory didn\u2019t simultaneously send every one of the product\u2019s hundreds of millions of users into crisis. And for some folks, the things that their chatbot might be fixated on are decidedly lower stakes than their ongoing divorce: one British immigrant to the US told the WSJ that ChatGPT kept trying to send him to British-style pubs against his will.<\/p>\n Still, experts have warned that for many consumers, the impact of expanded memory \u2014\u00a0especially for folks who spend a lot of time with chatbots \u2014 might be more subtle. After all, when engaging with a tool as hyper-personalized as memory-enabled ChatGPT can almost be like engaging with an AI-bottled version of your own mind. And as the University of Exeter philosopher and researcher Lucy Osler told the WSJ, that degree of hyper-personalization could work to \u201cconfirm certain self-narratives\u201d and \u201cmake them sound more real.\u201d <\/p>\n \u201cThey can box you in,\u201d said Osler.<\/p>\n Indeed, people may do well to remember that speaking to a chatbot, especially one with persistent memory, is often less like chatting with friend or neutral arbiter \u2014\u00a0and more like talking to a hall of mirrors.<\/p>\n Memory \u201ctakes people into cul-de-sacs,\u201d the ex-husband reflected. \u201cI would say [my wife] is like a Waymo driving around in a circle endlessly.\u201d<\/p>\n