Skip to main content

Cuban says Facebook knowingly runs AI deepfake scam ads

9 NOVEMBER 2026·2 MIN READ·5 SOURCES

Mark Cuban says Facebook knowingly runs scam ads made with AI-generated deepfakes. He says he has seen his own face in ads promoting nonexistent diabetes cures.

Cuban says Facebook knowingly runs AI deepfake scam ads

Key takeaways · 4

  • 01

    Cuban says he saw his face in AI-generated scam ads promoting nonexistent diabetes cures.

  • 02

    Meta projected $16 billion from scam ads; a separate projection put scams and banned-goods ads at 10% of 2024 revenue.

  • 03

    The reported scam-ad categories included illegal online casinos, fraudulent e-commerce and investment schemes, and banned medical products.

  • 04

    Senators sought an investigation and penalties, while Meta denied their allegations and said scam reports had fallen by 58%.

Cuban describes deepfake ads

Mark Cuban says Facebook knowingly runs AI-generated deepfake scam ads.[1] He says he has been seeing his own face in scammy ads.[1] The ads he described promoted diabetes cures that did not exist.[1] Separately, Gadget Review reported that older adults paid $294 for a diabetes cure Cuban had not endorsed.[4] These accounts concern deceptive promotions using a public figure’s likeness; the evidence provided does not establish how many ads appeared or how widely they circulated.

The figures are projections

Meta projected that scams and ads for banned goods would account for 10% of its 2024 revenue.[2] The Verge reported that Meta projected it would earn $16 billion from scam ads.[3] These are projections, not figures described in the evidence as realized revenue. The categories in the reported scam-ad estimates included illegal online casinos, fraudulent e-commerce and investment schemes, and banned medical products.[3] The 10% figure covers scams and banned-goods ads together, while the $16 billion figure is specifically described as projected scam-ad revenue.[2][3]

Reported responses to suspected scammers

Reuters reported that Meta’s responses to suspected rogue marketers included charging them a premium for ads.[2] Reports were also issued under the label “Scammiest Scammers.”[2] The evidence does not specify how many marketers were affected, how often premium charges were used, or whether those steps removed ads. For teams assessing platform controls, the distinction between a reported response and a demonstrated outcome matters: the available information describes the former, not a measured reduction attributable to those measures.

Senators request an investigation

Senators asked the FTC and SEC to investigate Meta’s ad practices and sought penalties as part of that request.[5] They cited fake ads impersonating U.S. government officials.[5] Meta denied the senators’ allegations and claimed scam reports had fallen by 58%.[5] The evidence does not establish the outcome of the requested investigation or independently verify Meta’s reported decline. Readers should therefore distinguish the senators’ claims, Meta’s response and the revenue projections rather than treating them as adjudicated findings.

For professionals responsible for advertising, compliance or platform decisions, the story raises questions about how deceptive ads are detected and what evidence supports claims about enforcement. Keep projections, allegations and a company’s account of falling reports distinct when evaluating risk or discussing the issue.

Why it matters
Daily session

Bring this to your team — one session a day, built for your industry.

kju.ai is built for teams: daily sessions, team analytics, and rollout support for your organization.

For teams

How this developed

  1. 9 November 2026

    Cuban says Facebook knowingly runs AI deepfake scam ads

Sources

AI fluency, one session a day, built for your work.