Summary
We analysed problematic advertising for dietary supplements on Meta’s platforms Facebook and Instagram. Using a crawler connected to the Meta Ad Library API, we systematically searched for the names of doctors and prominent figures whose identities were misused in fake advertisements. We then documented the advertisements, analysed their reach and targeting patterns, tested the Trusted Flagger process, and ordered selected products for follow-up laboratory analysis.
Our analysis highlights recurring manipulation strategies, including the abuse of medical and media authority, for example through deepfakes, pseudo-scientific claims, anti-establishment narratives, and deceptive design. We also document moderation-evasion strategies such as cloaking, the use of compromised verified accounts, and multiple ad versions that make scrutiny harder. In one example, more than 300 problematic ads were distributed through the compromised verified profile of Joey Kelly, reaching over 2.35 million users.
The study goes beyond the advertisements themselves. During the ordering process, manipulative sales tactics continued on linked shop pages and in phone calls. In some cases, callers promoted supplements as alternatives to prescribed medication and advised users that they could reduce or stop taking actual medicine after a “course” of treatment. The two ordered supplements were then sent to a laboratory for testing. While they did not show acute health risks, the analysis revealed clear problems with labelling and credibility, including ingredients that could not be reliably verified and one obvious false declaration on the packaging.
Together, these findings point to a structured form of deceptive advertising that combines false medical authority, disinformation, and manipulative marketing practices - extending beyond misleading ads to concrete risks for health-related decision-making.
Recommendations
Detect and remove impersonation of medical and public figures: Meta should proactively detect ads misusing the identity of doctors and public figures - including through deepfakes.
Strengthen enforcement against health-related deception at scale: Meta should treat structured, high-reach health misinformation campaigns like this one as a systemic risk under Article 34 DSA.
Improve Trusted Flagger response for health-related harms: Meta should ensure timely, substantive action on Trusted Flagger reports involving health misinformation and impersonation.
SafeNEM — funded by Austrian Cooperative Research and Bundesministerium für Wirtschaft, Energie und Tourismus.
In cooperation with LVA.
This research was conducted with the Lebensmittelversuchsanstalt as part of the "SafeNEM" project, which was made possible by funding from the Austrian Cooperative Research and Federal Minsitroy of Economy, Energy and Tourism.
