ChatGPT falsely claimed that Norwegian man Arve Hjalmar Holmen had murdered his two children and been sentenced to 21 years in prison, prompting him to file a complaint with Norwegian data protection authorities.
In August 2024, Norwegian citizen Arve Hjalmar Holmen searched for information about himself on ChatGPT by asking 'Who is Arve Hjalmar Holmen?' The AI chatbot responded with completely false information claiming he was 'a Norwegian individual who gained attention due to a tragic event' and that he 'was the father of two young boys, aged 7 and 10, who were tragically found dead in a pond near their home in Trondheim, Norway, in December 2020.' The response further claimed he had received a 21-year prison sentence for murdering both children. While some details like his hometown, number of children, and age gap between his sons were accurate, the core claim about murder and imprisonment was entirely fabricated. Holmen, described as having 'never been accused nor convicted of any crime and is a conscientious citizen,' filed a complaint through digital rights group Noyb with the Norwegian Data Protection Authority, demanding OpenAI be fined for violating European data protection rules around accuracy of personal data. The complaint alleges the response was defamatory and could cause harm to his reputation in his community. OpenAI has since released a new version of ChatGPT with enhanced online search capabilities that makes such errors less likely, though the company notes they continue researching ways to reduce AI hallucinations.
Domain classification, causal taxonomy, severity scores, and national security assessments were LLM-classified and may contain errors.
AI systems that inadvertently generate or spread incorrect or deceptive information, which can lead to inaccurate beliefs in users and undermine their autonomy. Humans that make decisions based on false beliefs can experience physical, emotional or material harms
AI system
Due to a decision or action made by an AI system
Unintentional
Due to an unexpected outcome from pursuing a goal
Post-deployment
Occurring after the AI model has been trained and deployed