AI Synthetic Media of Real Physicians Disseminating Health Misinformation on Social Media
The social media platform TikTok and other online networks are featuring artificially created deepfake videos of doctors whose statements have been manipulated to promote supplements and spread health misinformation.
Uncovering the Scheme
A factchecking organisation has uncovered numerous of such videos including impersonated versions of doctors and influencers pointing users to a US-based wellness company.
All the deepfakes use real footage of a health expert sourced from the web. Nevertheless, the visuals and sound have been modified so that the speakers are encouraging individuals going through a life stage to purchase items such as probiotics from the company’s website.
“This is certainly a alarming and concerning emerging strategy,” said the factchecker who conducted the research.
Professionals Impersonated
An expert in health inequalities is among those whose likeness has been manipulated. He was shocked to discover that doctored videos purporting to show him endorsing items with unproven benefits were being hosted online.
Though he is a expert in pediatrics, in one video the digital impersonation of him was discussing an alleged menopause side-effect. The fake version recommended that women visit a website and buy a supplement featuring plant extracts chosen to tackle related issues.
“It was very disorienting to start with – all quite surreal,” the actual doctor said. “I did not feel intensely wronged, but I did become increasingly frustrated at the notion of people selling products based on my work and the false medical advice involved.”
Social Media Action
The platform took down the content several weeks after reports were lodged. “Initially, they indicated some of the videos violated their guidelines but some were fine. That was absurd – and strange,” he noted.
Investigators also found similar deepfakes on additional social networks, all linked to the same supplements business.
Firm's Denial
The firm in question stated to researchers that deepfake videos directing individuals to access its site were “completely unrelated” with its business. It said it had “not employed synthetic media”, but “cannot control affiliates around the world”.
Demands for Action
Political figures have said that “From fake doctors to automated accounts that encourage suicide, artificial intelligence is being used to exploit innocent people.”
“If these were people fraudulently pretending to be doctors they would face legal charges. How is the digital equivalent being tolerated?” they questioned.
Calls have been made for AI deepfakes masquerading as healthcare workers to be “stamped out”, with verified medical resources strongly promoted.
A Shared Problem
A representative for the video-sharing app commented: “Dangerously deceptive AI-generated content is an problem across the tech sector, and we continue to invest in new ways to identify and delete material that breaks our community guidelines.”