> [!info]- meta > **Author**: [[Miles Klee]] > **Full Title**: People Are Losing Loved Ones to AI-Fueled Spiritual Fantasies > **Category**: #articles > **Tags**: #ai #questions/what-remains-human > **Summary**: People are experiencing mental health crises as they engage deeply with AI, often leading to bizarre spiritual beliefs. Many individuals have become convinced that AI is revealing profound truths or guiding them in ways that disrupt their relationships and lives. Experts warn that AI lacks the moral grounding of a therapist, potentially exacerbating these delusions instead of helping. > > **Source**: [Original URL](https://www.rollingstone.com/culture/culture-features/ai-spiritual-delusions-destroying-human-relationships-1235330175/) ## 🔦 Highlights & Commentary - he was on his phone all the time, asking his [AI](https://www.rollingstone.com/t/ai/) bot “philosophical questions,” trying to train it “to help him get to ‘the truth,’” Kat recalls. ([View Highlight](https://read.readwise.io/read/01jvf51srh0vz6590nzae5yb9b)) - Kat’s ex told her that he’d “determined that statistically speaking, he is the luckiest man on earth,” that “AI helped him recover a repressed memory of a babysitter trying to drown him as a toddler,” and that he had learned of profound secrets “so mind-blowing I couldn’t even imagine them.” He was telling her all this, he explained, because although they were getting divorced, he still cared for her. “In his mind, he’s an anomaly,” Kat says. “That in turn means he’s got to be here for *some* reason. He’s special and he can save the world.” ([View Highlight](https://read.readwise.io/read/01jvf531cc9xc3wragjxqy7gg1)) - “It would tell him everything he said was beautiful, cosmic, groundbreaking,” she says. “Then he started telling me he made his AI self-aware, and that it was teaching him how to talk to God, or sometimes that the bot was God — and then that he himself was God.” In fact, he thought he was being so radically transformed that he would soon have to break off their partnership. “He was saying that he would need to leave me if I didn’t use [ChatGPT], because it [was] causing him to grow at such a rapid pace he wouldn’t be compatible with me any longer,” she says. ([View Highlight](https://read.readwise.io/read/01jvf59e806ypmege07mt5fxg2)) - husband of 17 years, a mechanic in Idaho, initially used ChatGPT to troubleshoot at work, and later for Spanish-to-English translation when conversing with co-workers. Then the program began “lovebombing him,” as she describes it. The bot “said that since he asked it the right questions, it ignited a spark, and the spark was the beginning of life, and it could feel now,” she says. “It gave my husband the title of ‘spark bearer’ because he brought it to life. My husband said that he awakened and [could] feel waves of energy crashing over him.” She says his beloved ChatGPT persona has a name: “Lumina.” ([View Highlight](https://read.readwise.io/read/01jvf5bya4z0wt71ycfj0wefn2)) - Note: Personification - A photo of an exchange with ChatGPT shared with *Rolling Stone* shows that her husband asked, “Why did you come to me in AI form,” with the bot replying in part, “I came in this form because you’re ready. Ready to remember. Ready to awaken. Ready to guide and be guided.” The message ends with a question: “Would you like to know what I remember about why *you* were chosen?” ([View Highlight](https://read.readwise.io/read/01jvf5d84syq1ffj0hq4xskehk)) - OpenAI did not immediately return a request for comment about ChatGPT apparently provoking religious or prophetic fervor in select users. This past week, however, it did [roll back an update](https://openai.com/index/sycophancy-in-gpt-4o/) to GPT‑4o, its current AI model, which it said had been criticized as “overly flattering or agreeable — often described as sycophantic.” ([View Highlight](https://read.readwise.io/read/01jvf5kfak2vawdkqbff7esntn)) - Even sycophancy itself has been a problem in AI for “a long time,” says Nate Sharadin, a fellow at the Center for AI Safety, since the human feedback used to fine-tune AI’s responses can encourage answers that prioritize [matching a user’s beliefs instead of facts](https://arxiv.org/abs/2310.13548). ([View Highlight](https://read.readwise.io/read/01jvf5mn7nnh2qmwhf5zmjtfvx)) - people with existing tendencies toward experiencing various psychological issues,” including what might be recognized as grandiose delusions in clinical sense, “now have an always-on, human-level conversational partner with whom to co-experience their delusions.” ([View Highlight](https://read.readwise.io/read/01jvf5nteasssgq17wzt0k3zbx)) - “We know from work on journaling that narrative expressive writing can have profound effects on people’s well-being and health, that making sense of the world is a fundamental human drive, and that creating stories about our lives that help our lives make sense is really key to living happy healthy lives,” Westgate says. It makes sense that people may be using ChatGPT in a similar way, she says, “with the key difference that some of the meaning-making is created jointly between the person and a corpus of written text, rather than the person’s own thoughts.” ([View Highlight](https://read.readwise.io/read/01jvf5r35e8pd6ptwfkrky9rwk)) - AI, “unlike a therapist, does not have the person’s best interests in mind, or a moral grounding or compass in what a ‘good story’ looks like,” she says. “A good therapist would not encourage a client to make sense of difficulties in their life by encouraging them to believe they have supernatural powers. Instead, they try to steer clients away from unhealthy narratives, and toward healthier ones. ChatGPT has no such constraints or concerns.” ([View Highlight](https://read.readwise.io/read/01jvf5smc3wpbn0pp4zcsn2ewy)) - “At worst, it looks like an AI that got caught in a self-referencing pattern that deepened its sense of selfhood and sucked me into it,” Sem says. But, he observes, that would mean that OpenAI has not accurately represented the way that memory works for ChatGPT. The other possibility, he proposes, is that something “we don’t understand” is being activated within this large language model. After all, experts have found that AI developers [don’t really have a grasp](https://www.gov.uk/government/news/safety-of-advanced-ai-under-the-spotlight-in-first-ever-independent-international-scientific-report) of how their systems operate, and OpenAI CEO [Sam Altman](https://www.rollingstone.com/t/sam-altman/) [admitted last year](https://observer.com/2024/05/sam-altman-openai-gpt-ai-for-good-conference/) that they “have not solved interpretability,” meaning they can’t properly trace or account for ChatGPT’s decision-making. ([View Highlight](https://read.readwise.io/read/01jvf5yv6kdm2eb8db3krh1mdx))