ChatGPT Tells Woman Her Husband Was Cheating — Based On His Coffee Cup

A Greek woman has filed for divorce after ChatGPT allegedly interpreted coffee cup patterns and exposed her husband’s affair. The story, first reported by Greek City Times, blends the ancient art of tasseography with modern artificial intelligence, with dramatic consequences for a 12-year-old marriage.
<h3 data-start="443" data-end="474">AI Meets Ancient Ritual</h3>
Tasseography, the centuries-old practice of reading coffee grounds to divine one’s future, took an unconventional turn when the woman decided to seek answers from ChatGPT. She uploaded images of coffee cup remnants — hers and her husband’s — and asked the chatbot to interpret the symbols.
The response she claims to have received was enough to shatter the foundation of her marriage. According to her, ChatGPT warned that her husband was romantically involved with a younger woman, described as a "mysterious woman with the initial ‘E’," and that this person posed a threat to her family life. Her own cup, as per the AI, allegedly revealed signs of betrayal and looming domestic upheaval.
<h3 data-start="1166" data-end="1207">‘Then I Got a Call from a Lawyer’</h3>
The husband, who later appeared on a local television programme to share his side of the story, said he initially dismissed his wife’s claims as absurd.
“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,” he said.
Refusing any discussion about reconciliation or a mutual split, the woman formally served him divorce papers just three days after consulting the AI tool.
<h3 data-start="1739" data-end="1774">‘This Isn’t the First Time’</h3>
The husband also revealed that his wife has previously been influenced by pseudoscientific beliefs.
“A few years ago, she went to an astrologer, and it took her nearly a year to finally admit it was all nonsense,” he said.
His legal counsel has made it clear that the AI-generated reading holds no legal ground. “He is innocent until proven otherwise,” the lawyer said, pointing out that a chatbot’s interpretation of coffee grounds cannot be considered valid evidence of infidelity in court.
While AI continues to permeate our daily lives in practical ways — from healthcare to finance — this incident raises unsettling questions about its growing role in deeply personal and emotional decisions.