Facebook's AI Algorithms Promote Holocaust Denial Content Despite Ban

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.

Despite Facebook's announced ban on Holocaust denial content, its AI-driven recommendation algorithms continued to promote and network such material, leaving numerous denial pages active. This failure in AI enforcement facilitated the spread of anti-Semitic misinformation and hate speech, undermining the platform's stated policy and causing harm to affected communities.[AI generated]

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

The event involves Facebook's use of AI-driven recommendation algorithms that continue to suggest Holocaust denial content despite a stated policy ban. This algorithmic behavior directly facilitates the dissemination and network formation of harmful anti-Semitic content, which constitutes harm to communities and a violation of rights. Therefore, the AI system's use has directly led to harm, qualifying this as an AI Incident.[AI generated]
AI principles
AccountabilityFairnessSafetyRespect of human rightsTransparency & explainability

Industries
Media, social platforms, and marketing

Affected stakeholders
General public

Harm types
Human or fundamental rightsPublic interestPsychological

Business function:
Monitoring and quality control

AI system task:
Organisation/recommenders


Articles about this incident or hazard