Misinformation
AI systems can generate and spread false or misleading information, contributing to the pollution of the information ecosystem.
Risk Breakdown
Monthly Incidents
False or misleading information
Represents 60% of Misinformation risks
Examples:
- Generation of fake news articles
- Creation of misleading images or videos
- Hallucination of non-existent facts
Pollution of information ecosystem
Represents 40% of Misinformation risks
Examples:
- Amplification of conspiracy theories
- Filter bubbles and echo chambers
- Erosion of trust in authentic content
Related Incidents
AI-Generated Fake News
An AI text generator was used to create convincing but entirely fabricated news articles about a political candidate that went viral on social media.
Deepfake Scientific Research
AI-generated fake scientific papers with fabricated data and conclusions were submitted to academic journals, with several passing initial review stages.
Historical Fact Hallucination
A popular AI assistant confidently provided entirely fabricated historical events and dates that were later cited in student papers and online discussions.
Synthetic Media Confusion
AI-generated images of fictional events were mistaken for real photographs and used in news reporting without verification.
Mitigation Strategies
- Fact-checking integration
- Source attribution
- Content provenance techniques
- Media literacy education
- Transparent AI disclosures