A 21-year-old woman in South Korea allegedly used ChatGPT to help plan murders by asking questions about the fatal effects of mixing sleeping pills with alcohol, leading to two deaths and one attempted murder.
In January and February 2024, a 21-year-old woman identified as Kim in South Korea allegedly used OpenAI's ChatGPT to help plan a series of murders. Kim asked the chatbot questions such as 'What happens if you take sleeping pills with alcohol?' and 'Could it kill someone?' to determine if her cocktail would prove fatal. On January 28, Kim accompanied a man in his twenties to a Gangbuk motel in Seoul, gave him drinks laced with benzodiazepines she was prescribed for mental illness, and left alone two hours later. The man was found dead the following day. On February 9, Kim repeated the same process with another man in his twenties at another motel, who was also found dead. In December, Kim had attempted to kill a man she was dating by giving him a sedative-laced drink in a parking lot, but he survived after losing consciousness. Police initially arrested Kim on February 11 for inflicting bodily injury resulting in death, but upgraded charges after discovering her search history and ChatGPT conversations that established her alleged intent to kill. OpenAI confirmed that Kim's questions followed a factual line of questioning that would not raise alarms in the system's safety mechanisms, and the chatbot provided only factual responses.
Domain classification, causal taxonomy, severity scores, and national security assessments were LLM-classified and may contain errors.
Using AI systems to gain a personal advantage over others such as through cheating, fraud, scams, blackmail or targeted manipulation of beliefs or behavior. Examples include AI-facilitated plagiarism for research or education, impersonating a trusted or fake individual for illegitimate financial benefit, or creating humiliating or sexual imagery.
Human
Due to a decision or action made by humans
Intentional
Due to an expected outcome from pursuing a goal
Post-deployment
Occurring after the AI model has been trained and deployed