Danish man convicted of assault after using AI chatbot to plan attack
A 22-year-old man in Denmark used an AI chatbot to plan a violent attack on his former father-in-law, bypassing the chatbot's safety guardrails by falsely claiming he was researching for a book. He was convicted of aggravated assault and sentenced to ten months in prison after striking the victim twice in the head with a rubber hammer in February 2025.
What happened
A 22-year-old man struck his former father-in-law twice in the head with a rubber hammer in a parking lot in Ringsted, Denmark on February 5, 2025. The perpetrator planned the attack using an AI chatbot, asking it to research how to harm his victim without killing him.
Reported impact
The victim was physically injured in the attack. The motive was a problematic relationship between the perpetrator and the victim's daughter, the perpetrator's former wife.
Index analysis
The man bypassed the chatbot's safety guardrails by falsely claiming he was gathering knowledge for a book he wanted to write. According to the prosecutor, the incident demonstrates how general-purpose AI models can be manipulated into providing information on harmful topics.
Governance controls
Chatbot developers and deployers need to strengthen efforts to make their products less vulnerable to prompt engineering, according to analysis of the incident.
Classification
Sources and evidence
Sources not yet published
This record was researched and written by the Index from original reporting. Its sources are still being verified by an editor, and they are listed here once that check is complete. Until then the record carries no reviewer.