A marriage of 12 years has reportedly ended in dramatic fashion after a Greek woman allegedly took the “fortune-telling” interpretation of an artificial intelligence chatbot seriously, believing it had revealed her husband was cheating on her.
According to Greek City Times and reported by the Daily Mail, the unnamed mother-of-two submitted an image of her husband’s coffee cup remnants to ChatGPT, asking for a reading in the style of tasseography — an ancient form of divination that interprets patterns in tea leaves or coffee grounds. Traditionally performed by human fortune tellers, this method took a modern turn when the woman turned to AI for answers.
The bot allegedly told her that her husband was thinking about starting an affair with a woman whose name began with the letter “E” — and even went so far as to claim that he was already involved in an extramarital relationship. The woman reportedly took the prediction as truth.
Her husband, speaking on Greek morning television show To Proino, said he initially dismissed the claims as “nonsense.” However, the fallout was swift. “She asked me to leave, told our kids we were getting divorced, and then I got a call from a lawyer. That’s when I realised this wasn’t just a phase,” he said.
This isn’t the first time his wife has dabbled in the mystical. “A few years ago, she visited an astrologer and it took her a whole year to accept that none of it was real,” he added.
Within days of the AI reading, she filed for divorce. The husband is reportedly fighting back legally, arguing that the chatbot’s claims hold no evidentiary value. His lawyer insists there is no legal basis for the accusations and that the man remains “innocent until proven otherwise.”
The bizarre case has ignited a wave of reactions online, particularly on Reddit, where users mocked the situation as another example of AI overreach. “They’re taking out psychic jobs!” one commenter joked, while another quipped, “To be fair, they saw this coming.”
Others, however, expressed concern over the growing reliance on AI for personal decision-making, especially among emotionally vulnerable individuals. “We’re going to see a huge wave of people for whom LLM [Large Language Model] tools have just utterly annihilated their ability to comprehend reality,” one user warned.
While AI tools like ChatGPT are known for impressive language capabilities, they are not designed to interpret coffee grounds — nor are they trained in divination or spiritual insight. Yet, as this peculiar case demonstrates, some users may blur the lines between playful experimentation and serious life choices.
As legal proceedings unfold, the case serves as a surreal cautionary tale about the influence of AI in an increasingly tech-dependent world — and a stark reminder that even in the digital age, superstition hasn’t lost its hold.