Privacy advocacy group Noyb has filed a GDPR complaint against OpenAI after ChatGPT falsely claimed a Norwegian man murdered two of his children. As reported by Natasha Lomas for TechCrunch, the AI chatbot generated completely fabricated information stating that Arve Hjalmar Holmen was convicted and sentenced to 21 years in prison for killing his sons.
The case highlights serious concerns about AI hallucinations that present false personal information. While ChatGPT included some accurate details about Holmen, such as his hometown and that he has three children, it invented the horrific criminal allegations.
Noyb argues that OpenAI is violating the GDPR, which requires personal data to be accurate. “You can’t just spread false information and in the end add a small disclaimer saying that everything you said may just not be true,” stated Joakim Söderberg, a data protection lawyer at Noyb.
OpenAI responded that they continue researching ways to improve accuracy, noting this complaint relates to an older version of ChatGPT that has since been enhanced with search capabilities to reduce hallucinations.
Following a model update, the chatbot stopped producing the false claims about Holmen, but Noyb remains concerned that incorrect information could still exist within the AI model.
The complaint was filed with Norway’s data protection authority. GDPR violations can result in penalties of up to 4% of global annual turnover and potentially force changes to AI products.