‘Fake’ Doctors Encourage Patients to Skip Their Medication and Go for Mysterious Syrup
Misleading ads are spreading across social media platforms using images of well-known and respected doctors created by artificial intelligencepromoting “miracle” cures for diabetes or high blood pressure, a phenomenon that experts say is difficult to curb.
On Facebook and Instagram, these “fake” doctors promise to cure diabetes without the commonly prescribed metformin, even encouraging patients to ditch their medication, which they say will “kill them,” and opt for a mysterious “natural” drink instead.
These scams put the lives of many patients at risk, and are all the more dangerous because they are based on the images of famous doctors, such as Dr. Michel Sim in France or Dr. Michael Mosley in the United Kingdom, who are victims of the phenomenon of “Deep fakeThis is online.
“People trust these videos because these doctors have spent time building their credibility, so we believe them even when they make very outlandish claims,” British GP John Cormack, who is collaborating with the British Medical Journal (BMJ) on the subject, told AFP.
“These deep fakes really started this year,” says artificial intelligence specialist Henry Adger, noting that the ads are mainly targeting a “rather older” audience, by distorting the image of “doctors who appear a lot on television.”
Michel Sim, who was contacted by AFP in May, confirmed that he had filed a lawsuit against Meta regarding these “fraudulent operations.”
Briton Hilary Jones, who appears in AI-generated videos selling fake blood pressure medication and cannabis gum, has appointed an expert to monitor any similar ads.
“Even when these videos are taken down, they reappear the next day under different titles,” he says, via the British Medical Journal.
“Conspiracy theory circles”
Deepfake scams have “gone viral because of advances in deep learning,” says Frédéric Jory, a computer science teacher and researcher at the University of Caen.
He points out that the quality of deep fakes in images, audio clips and videos has advanced significantly, adding, “Today, we have tens of billions of images available to us, and we are able to build algorithms that can model everything that can appear in images and renew images, which is what we call generative artificial intelligence.”
In the latest phase, controversial figures such as Professor Didier Raoult in France have fallen victim to “deep fakes.”
Australian naturopath Barbara O’Neill, considered a “health hazard” by authorities in her country after promoting baking soda as an anti-cancer treatment, has become a TikTok star thanks to a “deep fake” that has seen her sell pills to “cleanse blood vessels”.
Contacted by Agence France-Presse, her husband, who co-manages the O’Neill company, expressed his regret that “many people use Barbara’s reputation to sell products that she does not personally recommend.”
This phenomenon does not surprise Henry Adger, who believes that “these people are models for some conspiracy theory circles.”
Among the fake videos featuring Barbara O’Neill are several conspiracy theories that claim the naturopath died, which is false, after discovering a miracle oil being sold on Amazon.
In the face of these “deepfakes,” experts are not very optimistic about the success of detection tools. “It’s a cat-and-mouse game,” says Gori. “In principle, detection can’t get much better. If someone finds a way to improve detection, someone who wants to create fake images will use the same method to circumvent them.”
The researcher highlights “technologies that ensure that the content has not been altered, as is the case with messages, thanks to programs that produce digital signatures, such as certificates.”
