Meta's AI Fact-Checking Labels Cause Harm, Deemed 'Opinion' in Court

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.

Meta (Facebook) used AI-supported fact-checking systems to label journalist John Stossel's content as misleading, reducing its visibility and causing reputational and economic harm. In court, Meta argued these fact-checks are merely 'opinions,' not factual statements, despite their significant impact on content distribution and user rights.[AI generated]

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

The article explicitly discusses Facebook's use of fact-checking systems that label content and reduce its distribution, which involves AI or algorithmic content moderation. The reporter alleges harm from these fact-checks, including defamation and censorship, which are violations of rights and cause economic harm. The AI system's use in content moderation is directly linked to these harms. Although Facebook claims the fact-checks are opinion, the practical effect is harm to the reporter. This fits the definition of an AI Incident because the AI system's use has directly led to harm to a person (reputational and economic).[AI generated]
AI principles
AccountabilityFairnessTransparency & explainabilityRespect of human rightsDemocracy & human autonomy

Industries
Media, social platforms, and marketing

Affected stakeholders
WorkersGeneral public

Harm types
ReputationalEconomic/PropertyHuman or fundamental rights

Business function:
Monitoring and quality control

AI system task:
Reasoning with knowledge structures/planning


Articles about this incident or hazard