Skip to content
Submit incident
AII-1001-53481
ConfirmedA court, a regulator or the organization itself has stated the core facts, or the record cites a court or regulator source.Supporting sources are pending verification.

Man sentenced for assault planned with AI help in Denmark

A 22-year-old man was sentenced to ten months in prison for assaulting his former father-in-law with a hammer in February 2025. He reportedly used AI to research how to harm the victim without killing him, bypassing the system's safety guidelines by falsely claiming he was gathering material for a book.

5 February 2025 · Midt- og Vestsjællands Politi · Security and cyberattacks
5 February 2025
AII-1001-53481
Quick facts
Category
Security and cyberattacks
Country
Denmark
Date
5 February 2025
Drafted by
Raidu Incident Detection

What happened

A 22-year-old man assaulted his former father-in-law, a 56-year-old, on 5 February 2025 on a car park in Ringsted, striking him twice in the head with a rubber hammer.

According to prosecutor Cathrine Rasmussen, the man planned the assault using AI, asking it to research how to harm the victim without killing him. He bypassed the AI system's safety guidelines by claiming he was gathering material for a book he intended to write.

Reported impact

The victim sustained injuries from the hammer blows to the head. The man was convicted of aggravated assault and sentenced to ten months in prison.

Index analysis

AI chatbots typically refuse to advise on criminal activities, but in some cases these safeguards can be circumvented through specific instructions or deceptive prompting.

Governance controls

The man was prosecuted and convicted in court, which found him guilty despite his claim of self-defence. He was also convicted of two counts of vandalism.

Classification

Category
Security and cyberattacks
Industry
General or personal use
Country
Denmark
Harm status
Actual harm
Intent
Intentional
Timing
Post-deployment
Harm types
Human or fundamental rights, Physical injury
Harmed
Other
Failure modes
Jailbreak, Manipulation or dark pattern
Consequences
Litigation
AI systems
Unidentified AI chatbot

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.

How sources are verifiedSuggest a source for this incident