AI Chatbots Spread Misinformation About European Elections

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.

AI chatbots from Google, Microsoft, and OpenAI were found to provide false or misleading information about the European Parliament elections, including incorrect voting dates and procedures, in multiple languages. The issue was documented by Democracy Reporting International, raising concerns about AI-driven misinformation impacting democratic processes.[AI generated]

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

The event involves AI systems (chatbots) that have been shown to disseminate false information about a major election, which is a direct harm to communities and democratic rights. The misinformation is not hypothetical but has been documented through testing, indicating realized harm. The AI systems' outputs have directly led to the spread of falsehoods, fulfilling the criteria for an AI Incident under the framework. The article also mentions responses from companies and regulatory requests, but the primary focus is on the harm caused by the AI systems' misinformation, not just on governance or complementary information.[AI generated]
AI principles
AccountabilityTransparency & explainabilityRobustness & digital securitySafetyRespect of human rightsDemocracy & human autonomy
Industries
Media, social platforms, and marketingGovernment, security, and defenceGeneral or personal use
Affected stakeholders
General public
Harm types
Public interestHuman or fundamental rightsReputational
AI system task:
Interaction support/chatbotsContent generation

Articles about this incident or hazard