This is a research prototype. The data and analyses are preliminary and not yet validated — we'd welcome your .
A patient named Beth Holland was defrauded of $300 after purchasing an ineffective lipedema cream advertised on Facebook using deepfake videos. The advertisements falsely featured endorsements from medical professionals and celebrities, including her own doctor. Investigations reveal a growing trend of AI-generated medical scams that exploit vulnerable patients, with experts warning of significant financial and health risks.
A U.S. patient, Beth Holland, reported losing money after purchasing a purported lipedema treatment, Svelta Venastra, promoted through an online advertisement that allegedly used deepfake video to falsely depict endorsements by medical professionals and public figures, including her own physician, Dr. David Amron. The reported ad circulated on Facebook and misrepresented the product's efficacy, leading Holland to spend approximately $300 on a cream that did not alleviate her condition.
Domain classification, causal taxonomy, severity scores, and national security assessments were LLM-classified and may contain errors.
Using AI systems to gain a personal advantage over others such as through cheating, fraud, scams, blackmail or targeted manipulation of beliefs or behavior. Examples include AI-facilitated plagiarism for research or education, impersonating a trusted or fake individual for illegitimate financial benefit, or creating humiliating or sexual imagery.
Also classified as
The deepfake ads exposed social media users to harmful, deceptive, and unapproved medical treatments.
The incident involves scammers using AI-generated deepfakes of celebrities and doctors to run fraudulent medical advertisement campaigns, targeting and deceiving individuals for financial gain.
Human
Due to a decision or action made by humans
Intentional
Due to an expected outcome from pursuing a goal
Post-deployment
Occurring after the AI model has been trained and deployed
The incident was caused by human scammers who intentionally used deployed deepfake AI tools to create deceptive advertisements and defraud consumers.
Added categories peak at 4: Severe (Epistemic Harm, Inferred)
An online consumer fraud incident where scammers used AI-generated deepfakes of medical professionals and celebrities to sell unapproved treatments. While highlighting the growing challenge of synthetic media and epistemic degradation, the incident is a cybercrime issue with negligible direct national security implications.
The AI systems (deepfake generators) were used as tools by human scammers to perform a specific function—generating synthetic video and voice—based on direct inputs (pirated footage) without autonomous planning or decision-making.
Limited Risk: The report describes the use of AI-generated content (deepfakes), which falls under Risk Level 3 due to transparency obligations to inform users they are interacting with synthetic media.
The AI was used to digitally manipulate the images and voices of doctors and celebrities to create realistic, fake endorsement videos.
The report provides clear, detailed accounts of the scam mechanism, the specific victims, the financial loss of $300, and the types of deepfakes used. The role of AI as a tool for generating deepfakes is explicitly stated.
None. The incident is clearly a financial scam leveraging deepfake technology.
Technology
Advanced Deepfake Generation
Highly realistic AI-generated voices and likenesses are easy to create.
Difficult Detection
AI technology is so advanced that deepfakes are nearly impossible to spot.
Data Inputs
Pirated Media Footage
Scammers easily harvest real doctor and celebrity footage from the internet.
Fake Compliance Certificates
Scammers input fabricated FDA certificates to bypass verification systems.
Human Factors
Desperation for Relief
Patients in pain are highly susceptible to promising medical scam ads.
Trust in Authority Figures
Victims trust deepfaked endorsements from recognized doctors and celebrities.
Process and Methods
Inadequate Ad Moderation
Social media platforms fail to proactively block fraudulent deepfake ads.
Ineffective Takedown Processes
Takedowns occur only after scams are reported, allowing victims to be harmed.
Management
Inadequate Risk Assessment
Platforms fail to anticipate and mitigate risks of AI-driven medical fraud.
Prioritizing Ad Revenue
Social media platforms prioritize ad volume over rigorous safety verification.
Regulatory Environment
Cross-Border Jurisdictions
Scammers operate globally through shell companies, hindering law enforcement.
Lack of Regulatory Oversight
Absence of strict regulations or enforcement on AI-generated medical ads.
AI system involved
An AI system was involved, but the reports do not identify it.
Identified by automated extraction; can differ from the alleged developers above.
