
The information displayed in the AIM should not be reported as representing the official views of the OECD or of its member countries.
Doubao, a Chinese AI assistant, mistakenly displayed an altered image of actor Fan Wei instead of historical figure Li Yuanhong due to widespread online misinformation and flawed training data. The incident exposed AI's vulnerability to 'hallucinations,' leading to public concern over AI reliability and the spread of false information. The issue was later corrected.[AI generated]
Why's our monitor labelling this an incident or hazard?
The AI system 'Doubao' was used to retrieve images of a historical figure but mistakenly showed incorrect images due to misidentification caused by similar appearances and widespread online image misuse. This error led to misinformation, which is a form of harm to communities. Since the AI system's malfunction directly caused this misinformation, it qualifies as an AI Incident. The article also includes discussion about the AI tool's monetization and responsibility for accuracy, but the primary event is the realized harm from the AI's erroneous output.[AI generated]