TikTok's recommendation system spreads AI-generated fake ads and health misinformation
AI-generated fake advertisements for fraudulent health products, including fake weight loss supplements and cloned pharmacy websites, are spreading across TikTok despite the company's stated anti-misinformation policies. TikTok's recommendation algorithm reportedly makes health-related content "stickier," keeping it in users' feeds longer, while a 2022 study found over half of ADHD-related videos on the platform were misleading.
What happened
AI-generated advertisements promoting fraudulent health products are proliferating on TikTok, including fake GLP-1 weight loss patches and unproven cortisol belly cures, many linking to cloned websites impersonating legitimate Irish pharmacies.
The Irish Examiner reports that despite TikTok's stated policies prohibiting misleading and deceptive advertising, these ads are slipping through weak content filters. The Washington Post's investigation found that TikTok's recommendation system treats mental health topics as "stickier" than other subjects, making them linger in users' feeds longer.
Reported impact
A 2022 study found that 52% of TikTok videos about ADHD could be classified as misleading, with only 21% considered useful and 27% sharing personal experience.
Users report difficulty removing mental health-related content from their feeds even when expressing disinterest, exposing them to misinformation about medical conditions and fraudulent health products.
Index analysis
TikTok's parent company ByteDance invested $12 billion in AI chips in 2025, coinciding with a surge in AI-generated content on the platform. The company acknowledged the investigation but disputed that the reporting reflected how its recommendation system works.
Governance controls
TikTok's stated policies prohibit behaviours that are misleading, inauthentic and deceptive, warning that accounts engaging in such practices risk suspension or banning. However, enforcement of these policies appears insufficient to prevent the spread of fraudulent ads and health misinformation.
Classification
Sources and evidence
This record was researched and written by the Index, based on the following original reporting.