A mother of two from Greece reportedly took an AI chatbot’s fortune-telling advice seriously after it claimed her husband was cheating on her during a coffee break with a colleague.
According to a Greek news outlet cited by the Daily Mail, the woman—who has been married for 12 years—asked the chatbot to interpret the leftover coffee grounds in her husband’s cup and provide a reading.
This unusual request mirrors the ancient practice of tasseography, where patterns in tea leaves, coffee grounds, or wine sediments are interpreted to predict the future—traditionally done by fortune tellers, not AI.
However, the woman’s experiment with the digital oracle reportedly led to a dramatic marital crisis when the chatbot suggested infidelity.
Greek City Times reported that she shared a photo of the coffee cup’s bottom with the chatbot, which produced a surprising response that shook both her and her husband.
Read Also: OpenAI finalizes $40 billion funding at $300 billion value
The husband later appeared on the Greek morning show To Proino, explaining that this was not the first time his wife had turned to supernatural methods for guidance.
“A few years ago, she visited an astrologer and it took her a whole year to accept that none of it was real,” he told the morning show.
However, the reading she received from ChatGPT appeared to have influenced her entirely.
After she uploaded photos of her and her husband’s coffee cups into the AI chatbot, she was allegedly told that her partner was thinking about starting an extramarital affair with a woman whose name began with the letter E.
ChatGPT allegedly went even further with the reading and apparently interpreted the coffee grounds to suggest that he was already cheating on her and the ‘other woman’ had plans to tear their marriage apart.
Speaking on To Proino, the husband said, “I laughed it off as nonsense, but she took it seriously.
“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.”
According to the publication, he refused to agree to a mutual separation, but only three days later, he was formally served with divorce papers.
He is reportedly retaliating against his wife’s divorce attempts, with his lawyer arguing that Chatgpt’s claims have no legal standing and that he is ‘innocent until proven otherwise’.
Many people have responded to the bizarre story on Reddit, with some people joking that AI is now stealing a new genre of work.
“They are taking out psychic jobs!’ one person wrote, to which another replied with a pun: ‘To be fair, they saw this coming.”
A third pointed out that, while developments in AI have been impressive, the technology is flawed and can come up with some ‘extremely stupid’ results at times.
“I was running a bunch of things through ChatGPT to get some quick examples set up for a statistics course and it insisted that the word ‘extraterrestrial’ has 15 letters, of which four of them were E’s,” they gave as an example.
“No matter how I worded the question it insisted both of these things. Now I can’t recreate that, which is even more confusing.”
On a more serious note, one person pointed out that vulnerable people are finding it harder to tell the difference between reality and fiction thanks to tools like AI chatbots.
“I genuinely feel like we’re gonna see a huge wave of people for whom LLM (Large Language Model) tools have just utterly annihilated their ability to comprehend reality,” they wrote.
“I’d say it’s a technology that needs safeguards but that ship sailed years ago.”
