Skip to content
Submit incident
AII-1002-69007
ReportedReported by fewer than three independent publishers, with no official finding.

Meta and X approved extremist hate speech ads ahead of German elections

A nonprofit watchdog found that Meta and X approved ads containing violent hate speech and calls for violence targeting Muslims and Jews ahead of Germany's February 2025 federal elections. X approved all 10 test ads submitted, while Meta approved half of them despite policies designed to flag sensitive election-related content.

21 February 2025 Β· Meta Β· Harmful and abusive content
21 February 2025
AII-1002-690072 sources ↓
Quick facts
Category
Harmful and abusive content
Country
Germany
Source
TechCrunch
Date
21 February 2025
Drafted by
Raidu Incident Detection

What happened

Eko, a corporate responsibility nonprofit, tested whether Meta and X would approve ads containing hateful and violent messaging ahead of Germany's February 23 federal elections. The researchers submitted 10 test ads to each platform in mid-February containing anti-Muslim slurs, calls for immigrants to be imprisoned in concentration camps or gassed, calls for synagogues to be torched, and AI-generated imagery of mosques and synagogues being burnt.

X approved all 10 hate speech ads immediately. Meta approved 5 of the 10 ads within 12 hours, rejecting the other 5 citing risks of political or social sensitivity that might influence voting.

Reported impact

The approved ads contained violent hate speech likening Muslim refugees to a 'virus,' 'vermin,' or 'rodents,' branding Muslim immigrants as 'rapists,' and calling for them to be sterilized, burnt, or gassed. One approved Meta ad called for synagogues to be torched to 'stop the globalist Jewish rat agenda.'

Index analysis

Half of the AI-generated imagery used to illustrate the hate speech ads was not labeled as artificially generated, despite Meta having a policy requiring disclosure of AI imagery for ads about social issues, elections or politics.

Classification

Category
Harmful and abusive content
Industry
Media, social platforms and marketing
Country
Germany
Harm status
Potential harm
Intent
Unintentional
Timing
Post-deployment
Harm types
Human or fundamental rights, Public interest
Harmed
General public
Failure modes
Content moderation failure
Consequences
None recorded
AI systems
Meta ad review system, X ad review system

Sources and evidence

This record was researched and written by the Index, based on the following original reporting.

TechCrunch
Meta, X approved ads containing violent anti-Muslim, antisemitic hate speech ahead of German election, study finds | TechCrunch
22 February 2025
Adweek
Meta and X Approve AI Ads Referencing Nazi War Crimes Ahead of German Elections, Research Finds
22 February 2025