Skip to content
Submit incident
AII-1002-61229
ConfirmedA court, a regulator or the organization itself has stated the core facts, or the record cites a court or regulator source.

Brazilian Government Demands Removal of AI-Generated Fake Video of Finance Minister

Brazil's government demanded removal of an AI-generated deepfake video showing Finance Minister Fernando Haddad falsely announcing nonexistent taxes on social media. Meta removed the content after an official notice, and the video now displays a synthetic media warning on Instagram.

9 January 2025 · Meta · Misinformation and synthetic media
9 January 2025
AII-1002-612292 sources ↓
Quick facts
Category
Misinformation and synthetic media
Country
Brazil
Source
CNN Brasil
Date
9 January 2025
Drafted by
Raidu Incident Detection

What happened

A video manipulated using artificial intelligence circulated on Meta platforms showing Brazil's Finance Minister Fernando Haddad making false statements about new taxes on pet ownership, prenatal services, and sports betting prizes.

The Advocacia-Geral da União (AGU) sent an extrajudicial notice to Meta on January 9, 2025, demanding removal of the video within 24 hours, arguing it contained fraudulent information with false attributions to the minister and violated the platform's terms of use and community standards regarding interference with political processes.

The AGU's analysis identified telltale signs of artificial manipulation including abrupt cuts, visible changes in lip movement, and discrepancies in voice timbre typical of generative AI-created content.

Reported impact

The fake video was designed to confuse the public about the Finance Minister's actual positions on matters of public interest regarding taxation policy.

Governance controls

Meta removed the video from its platforms following the AGU's notification, and the content on Instagram now displays a warning indicating it was generated by artificial intelligence.

Classification

Category
Misinformation and synthetic media
Industry
Media, social platforms and marketing
Country
Brazil
Harm status
Potential harm
Intent
Unknown
Timing
Unknown
Harm types
Reputational, Public interest
Harmed
General public, Government
Failure modes
Deepfake video
Consequences
Content removal
AI systems
Generative AI deepfake

Sources and evidence

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

CNN Brasil
Governo pede que Facebook remova em até 24 h vídeo de Haddad adulterado com IA
9 January 2025
odia.ig.com.br
Brazilian Government Demands Removal of AI-Generated Fake Video of Finance Minister