Sydney Man Arrested After AI Chatbot Flags Mass Shooting Threat

Thumbnail Image

The information displayed in the AIM should not be reported as representing the official views of the OECD or of its member countries.

David Lan Thackray, 29, was arrested in Sydney after allegedly telling the AI chatbot Claude about plans to carry out a mass shooting at a childcare centre. The AI system flagged the threat, leading to police intervention and charges of using a carriage service to menace, harass, or offend.[AI generated]

Why's our monitor labelling this an incident or hazard?

The AI system (Claude chatbot) was used by the individual to communicate a threat of mass violence, which is a direct potential harm to public safety (harm to health of groups of people). The AI system's safety features detected and flagged the threat, leading to law enforcement intervention. The event involves the use of an AI system and its outputs directly linked to a serious potential harm scenario. Although the harm has not yet materialized, the AI system's role in the chain of events is pivotal and directly related to the threat. This fits the definition of an AI Incident because the AI system's use has directly led to a significant harm-related event (arrest and legal proceedings related to a threat of mass violence).[AI generated]
Industries
Consumer services

AI system task:
Interaction support/chatbotsEvent/anomaly detection


Articles about this incident or hazard