Fake AI doctors are racking up millions of TikTok views with bogus health advice
Computer-generated medics look convincing enough to fool viewers. Doctors are alarmed.
Key points
- AI-generated doctor personas on TikTok are gaining millions of views spreading dubious health claims.
- The British Medical Association has flagged these accounts as a "huge danger to public safety".
- Experts warn the accounts promote "medical myths" and fake miracle cures to ordinary viewers.
- There is currently no platform-wide system that reliably labels AI-generated health content.
A researcher pulls up a TikTok video. On screen, a confident-looking doctor explains a health remedy in clear, calm language. The account has millions of views. The doctor does not exist.
These are AI-generated personas, digital characters built with artificial intelligence tools that can produce realistic-looking human faces, voices and mannerisms from scratch. They look like real people. They sound authoritative. And according to new research, first flagged by The Guardian, they are spreading medical advice that experts say is flat-out wrong.
Why does this matter to ordinary people?
Because most viewers cannot tell the difference, and the advice can cause real harm.
Dr Emma Runswick, deputy chair of the British Medical Association council (the main professional body representing doctors across the UK), warned that these accounts "peddle medical myths and promote so-called miracle cures." Her words were direct: misleading health claims online pose a "huge danger to public safety."
Think of someone with a new diagnosis scrolling TikTok late at night. A polished video appears, featuring what looks like a doctor, offering a simple fix. That viewer might delay seeing a real clinician, stop prescribed medication, or spend money on a product that does nothing.
That is the practical danger here.
How are these videos getting so many views?
TikTok's algorithm, the system that decides which videos to show each user, rewards content that keeps people watching. Health content, especially anything that promises an easy answer to a difficult problem, tends to perform well. AI tools now make it cheap and fast to produce polished-looking video at scale, meaning bad actors can flood the platform with dozens of clips in the time it once took to film one.
The faces are generated, the voices are synthesised, and the scripts are written or shaped by AI. None of it requires a real doctor, a real studio, or much money at all.
What should viewers actually do?
Treat any health advice on short-video platforms with real scepticism, whether or not the presenter looks human.
A few practical checks help. Search the account for a verifiable real-world identity: a name you can find on a medical register, a hospital profile, or a professional organisation. If there is none, that is a warning sign. Check whether the claimed treatment appears on a reputable health service website. When in doubt, ask an actual clinician before changing anything about your care.
Regulators and platforms have not yet caught up with how quickly AI-generated content spreads. Until they do, the burden of scepticism falls on the viewer.
Common questions
Can TikTok detect AI-generated health videos automatically?
Not reliably. TikTok has tools to label some AI-generated content, but they do not catch everything, and accounts actively trying to look real can slip through. The research cited here found millions of views on accounts that had not been flagged.
Is watching this kind of content illegal?
No, but creating and distributing deliberate medical misinformation may breach platform rules and, depending on the country and the specific claim, could attract regulatory attention. For now, enforcement is patchy.
