AI-generated anti-immigrant and sexualized content spreads on TikTok
Researchers found over 350 TikTok accounts posting billions of views of AI-generated content, including anti-immigrant material and sexualized imagery of minors, with inadequate labeling and moderation by the platform. The accounts used automated posting patterns to spread the content widely while evading detection for extended periods.
What happened
Researchers from AI Forensics uncovered 354 AI-focused accounts on TikTok that had accumulated 4.5 billion views over a month-long period in 2025, posting 43,000 pieces of AI-generated content.
Some of the accounts posted up to 70 times a day, with most accounts launched at the beginning of the year, indicating automated posting designed to game TikTok's recommendation algorithm. Content included anti-immigrant material, fake broadcast news segments with anti-immigrant narratives, and sexualized content featuring female bodies including girls that appeared to be underage.
AI Forensics found that half of the content posted by these accounts was not labeled as AI-generated, and less than 2% carried TikTok's official AI content label. Some accounts escaped TikTok's moderation for months despite posting content that violated the platform's terms of service.
Reported impact
The AI-generated content, including fake news segments using known broadcasting brands such as Sky News and ABC, reached billions of users on the platform and could mislead viewers about its origins.
Content sexualizing female bodies, including apparent minors, was widely distributed on the platform without proper moderation.
Index analysis
The incident reflects gaps in TikTok's content moderation systems, which failed to consistently catch and remove violative content despite it being posted repeatedly from automated accounts over extended periods.
The low rate of AI labeling on the platformβless than 2% in these accounts despite TikTok's availability of an AI labelβenabled deceptive presentation of generated content as authentic material.
Governance controls
TikTok said the report's claims were "unsubstantiated" and stated the researchers had singled it out for an issue affecting multiple platforms. The platform offers users an option to reduce the amount of AI content they see.
Some of the posts were taken down by TikTok after being referred to the platform by the Guardian, and dozens of the accounts identified in the study were subsequently deleted, indicating moderator action.
Classification
Sources and evidence
This record was researched and written by the Index, based on the following original reporting.