What evidence exists of deepfake endorsements being used in supplement marketing campaigns?
This fact-check may be outdated. Consider refreshing it to get the most current information.
Executive summary
Documented evidence shows that bad actors have used AI-generated deepfakes to create fake endorsements for dietary supplements and other health products: fact‑checking groups and multiple news outlets have found hundreds of clips that repurpose real clinicians’ footage to make it appear they endorse specific supplements, and investigators have traced some campaigns to affiliate‑style funnels selling unproven products . Technology vendors and consumer‑protection reporting also document celebrity and doctor impersonations in paid ads and organic posts, while platform takedowns and legal commentary reveal enforcement and regulatory gaps [1].
1. Documented incidents: what investigators have found
UK fact‑checking charity Full Fact and mainstream outlets reported “hundreds” of AI‑generated clips that reuse or manipulate conference and broadcast footage of clinicians to make it appear they endorse supplements sold through affiliates, with specific networks promoting menopause and weight‑loss supplements cited in their investigations . Independent analyses and marketing writeups have catalogued named examples — including deepfakes featuring public figures such as Rebel Wilson and alleged clips using clinicians like Dr. Robert Lustig as a target — that directed viewers to product pages for items like “Keto Flow + ACV,” Glyco Balance, and other unproven remedies .
2. Methods and patterns used in campaigns
Reporting and vendor research describe a familiar playbook: reuse of authentic footage of experts, AI‑based voice and lip‑sync manipulation to change the message, and deployment in short‑form video ads or influencer‑style posts that lead to landing pages full of fake testimonials and affiliate links — a funnel designed to convert trust into sales [1]. Analysts note the use of multilingual deepfakes and demographic targeting to place these videos where local trust in particular clinicians or TV personalities is highest, magnifying credibility [1].
3. Platforms, reach and persistence
Social platforms including TikTok, Instagram, Facebook, YouTube and even search ad placements have hosted these clips, sometimes for days or weeks before removal; some high‑view deepfakes accumulated hundreds of thousands of views before takedown, demonstrating both reach and uneven enforcement by platforms . Industry observers and fact‑checkers warn that cheap, widely available tools make creating passable celebrity or expert fakes easy for tech‑savvy scammers, increasing volume and making detection harder .
4. Motives, attribution and the role of affiliate marketing
Evidence in reporting links these deepfake endorsements not to respected brands but to third‑party affiliate funnels and dubious supplement vendors that deny direct involvement when exposed; the commercial motive is clear — drive purchases of unproven or unsafe products by harnessing fabricated authority . Attribution to specific operators is often difficult in public reporting: investigators frequently identify the downstream sellers or affiliate domains but not necessarily the creators of the synthetic media, leaving gaps in who profits and how campaigns are organized .
5. Legal, regulatory and platform response — active but limited
Legal scholars note that existing publicity, trademark and false‑advertising laws provide tools to challenge unauthorized commercial use of a person’s likeness, and regulators require clear disclosure for endorsements, but enforcement against synthetic‑media scams has lagged and platform takedowns are inconsistent . Media outlets cite platform removals after complaints and calls from medical bodies for better protections, but also document that some deepfakes remained online and that rules specific to deepfakes were sparse at the time of reporting .
6. Assessment, open gaps and what evidence does not yet show
The existing body of reporting and fact‑checking provides strong, concrete examples that deepfake endorsements are being used in supplement marketing campaigns and that the technique is widespread enough to produce hundreds of clips and major view counts . What public sources do not consistently show is systematic attribution to named corporate perpetrators behind every campaign or definitive industry‑scale metrics quantifying total global losses specific to deepfake‑driven supplement sales; those remain investigative gaps for reporters and regulators .