After the scandal surrounding Grok, the rise of deepfakes does not seem to have finished disrupting the digital ecosystem. Indeed, like other AI technologies, deepfakes have also become a fertile ground for online scams. Faced with the proliferation of frauds on its platforms, Meta has decided to take legal action.
In an official announcement, it was revealed that the parent company of Facebook and Instagram has filed several complaints against advertisers based in Brazil, China, and Vietnam. According to the group, advertising campaigns used deepfakes of celebrities to deceive internet users and promote fraudulent products.
The most common technique, known as “celeb-bait,” involves exploiting the image, and sometimes the voice, of public figures using hyper-realistic montages. Thanks to generative AI, fraudsters create fake videos where celebrities appear to recommend a product or service.
In Brazil, Meta is suing two companies and four individuals accused of orchestrating a fraud operation in the healthcare sector. The image of the doctor and oncologist Drauzio Varella was allegedly used without authorization to promote unregulated products. While the legal action is applauded, the doctor stated in the local press that “it is just a drop in the ocean” compared to the extent of public health-related scams.
In China, a company is accused of broadcasting similar ads targeting internet users in the U.S. and Japan, aiming to lure them into supposed investment groups actually designed to extract money from victims.
Meta claims that its protection program now covers over 500,000 public figures worldwide to limit the fraudulent use of their image.
Beyond deepfakes, the lawsuits also target another practice called “cloaking.” This technique involves concealing a legitimate version of a site from ad verification systems, while users are redirected to a fraudulent page.
An advertiser from Vietnam is specifically accused of offering fake Longchamp brand articles at discounted prices. Victims, enticed by these offers, shared their banking details for products that were never delivered, some even ending up enrolled in paid subscriptions.
Faced with these increasingly sophisticated methods, Meta states that it is strengthening its detection tools based on artificial intelligence to identify suspicious campaigns more quickly. This technological response to a threat, also fueled by AI, illustrates the growing complexity of the fight against online fraud.
