Mark Cuban has publicly accused Meta of knowingly running AI-generated deepfake scam ads that use his face and voice to sell bogus health products, sharpening a debate over how much the company earns from fraudulent advertising and how little it does to stop it. The billionaire investor and co-founder of Cost Plus Drugs says the ads are still live on Facebook despite the platform's automated systems flagging them.
Cuban laid out the accusation on an August episode of Pivot, the podcast hosted by Kara Swisher and Scott Galloway, and his comments have circulated widely this week as scrutiny of Meta's advertising business intensifies. "Every single week I get emails from little old men and ladies saying, 'Will my cure for diabetes work?' or 'I just ordered your cure for diabetes for $294,'" Cuban said. For more context on this story, see our coverage of latest AI developments.
The Deepfakes Targeting Cuban Himself
The ads, Cuban explained, are built with an AI copy of his face and voice pitching a fake diabetes cure. The irony is pointed: Cuban genuinely co-founded Cost Plus Drugs, an online pharmacy that sells low-cost generic medications, which gives the scam a veneer of credibility it never earned.
"Facebook knows the little old ladies are getting ripped off," Cuban said, calling the platform "the worst of the worst." He noted that some of the deepfake ads even include a disclaimer admitting the person on screen is AI-generated and the clip is meant as entertainment — but by the time a viewer reaches that text, the pitch has already done its work. "And to Facebook, that's good enough," he said.
"The point is, that defines the culture of the organization," Cuban added, arguing that a company willing to let scammers target people's grandparents has revealed what it will tolerate at every level of the business.
What Meta's Own Documents Reportedly Showed
The accusation lands against the backdrop of internal Meta documents first reported by Reuters on November 6, 2025. According to that reporting, Meta internally projected roughly $16 billion in revenue — about 10 percent of its 2024 total — from ads pushing scams and prohibited products.
The same documents indicated that users across Meta's apps were seeing some 15 billion "higher risk" ads per day, the category with obvious markers of fraud. One late-2024 internal document put the value of that category at about $7 billion a year.
The 95 Percent Threshold
Perhaps the most damaging detail from the documents concerns how Meta decides when to act. According to a May 2025 presentation prepared by Meta's own safety staff, the company's platforms play a part in roughly a third of the scams that succeed in the United States.
Before Meta bans an advertiser outright, its automated systems must judge the account at least 95 percent likely to be running a scam, the documents say. Below that threshold, a suspected scammer simply pays more per ad and stays live — a policy that, critics argue, effectively lets fraud continue as long as it pays.
Meta's Defense
Meta has pushed back hard on the characterization. Company spokesman Andy Stone told Reuters the documents "present a selective view that distorts Meta's approach to fraud and scams," and described the 10 percent revenue estimate as "rough and overly-inclusive."
Stone also pointed to the company's enforcement record: complaints from users about scam ads had fallen 58 percent over the prior 18 months, he said at the time, and Meta pulled down more than 134 million scam ads in 2025. The same internal documents, however, note that the Securities and Exchange Commission has been looking into Meta over advertising tied to financial fraud — a detail that suggests regulatory interest in exactly the questions Cuban is raising.
The Toll on Older Americans
The human cost behind these numbers is substantial. Older Americans lose more to this kind of crime than any other age group. People 60 and older told the FBI's Internet Crime Complaint Center they lost $7.7 billion in 2025, across 201,266 complaints — 59 percent more money than the year before, according to the bureau's 2025 Internet Crime Report.
Investment schemes did the worst damage at $3.5 billion, with tech and customer support scams taking another $1.04 billion. AI showed up in more than 3,100 of those complaints, and together those victims lost over $352 million — and that only counts the people who reported the crime.
Why the Accusation Caries Weight
Cuban is an unusual messenger for this critique. Most public figures whose likenesses are hijacked by deepfake scammers have no direct connection to the products being faked. Cuban does — and the scammers exploit the trust he built through a legitimate pharmacy to sell a fictitious one. That firsthand view of the harm, he argues, gives him a clearer picture than most of what Meta's ad-review systems let through.
His broader point is cultural rather than legal: a platform that tolerates this behavior has made a choice, and the choice is visible everywhere in the organization. Whether regulators agree — and whether the SEC's reported interest turns into formal action — will determine if that choice finally carries a price.
---
Stay Ahead of AIGet the latest AI news, analysis, and breakthroughs — all in one place.
Read more AI news →
