ChatGPT helped a woman harass her husband. Initially, the US resident used the chatbot for work, but then began discussing personal issues with it, including divorce. When the couple thought about it, she feared that her husband would deprive her of her money and her children. And ChatGPT only confirmed her fears. As a result, the American woman started asking the chatbot how she could poison a man. The censorship didn't work, and the AI told her about drugs, lethal doses, and ways to hide the crime. At one point my husband became very ill: he suffered from stomach pains for three days and could not think clearly. Then the man suspected something was wrong. Then he found records where the woman recorded the doses and contacted the police. Initially, a case was opened against her, but was later dropped due to lack of evidence. Now the American is trying to bring OpenAI to justice because ChatGPT participated in the attempt on his life.