Skip to content
Submit incident
AII-1000-78103
ReportedReported by fewer than three independent publishers, with no official finding.

AI-Generated Avatars Spread Pro-Trump Disinformation Ahead of US Midterms

Hundreds of AI-generated avatars posted pro-Trump disinformation on social media platforms ahead of the 2026 US midterm elections, using hyper-realistic images to spread political messaging. Trump himself shared content from an AI avatar making unfounded allegations against a political rival, and researchers note the technology is increasingly difficult to detect.

10 May 2026 Β· Donald Trump Β· Misinformation and synthetic media
10 May 2026
AII-1000-78103View source β†—
Quick facts
Category
Misinformation and synthetic media
Country
United States
Source
Economic Times
Date
10 May 2026
Drafted by
Raidu Incident Detection

What happened

Hundreds of AI-generated avatars have flooded social media platforms including TikTok, Instagram and Facebook posting pro-Trump political content ahead of the 2026 US midterm elections.

The AI avatars, depicted as blonde women in swimsuits or military fatigues and young men dressed as immigration agents, parrot Trump's slogans such as 'America First' and echo his talking points on immigration and other issues.

Earlier in 2026, Trump himself posted on Truth Social a video of a platinum blonde AI avatar pushing unfounded corruption allegations against California Governor Gavin Newsom.

Reported impact

The Governance and Responsible AI Lab at Purdue University has tracked numerous AI-generated pro-Trump accounts across multiple platforms, with researchers noting the technology can be 'increasingly difficult to detect, especially when made by sophisticated operators.'

Index analysis

It is not clear who is behind the AI accounts or whether they are part of a coordinated influence operation ahead of November elections that will determine whether Republicans maintain control of Congress.

Classification

Category
Misinformation and synthetic media
Industry
Media, social platforms and marketing
Country
United States
Harm status
Potential harm
Intent
Unknown
Timing
Unknown
Harm types
Public interest
Harmed
General public
Failure modes
Deepfake image, Manipulation or dark pattern
AI systems
AI-generated avatars

Sources and evidence

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

Economic Times
Fervent and fake: High-glam AI avatars boost Trump ahead of midterms - The Economic Times
10 May 2026