--- title: "Your founder's face in an ad you never bought" description: "A deepfake founder ad is removed by targeting the funnel: the advertiser account paying for delivery, the landing page and the domain, not just the video creative." date: "2026-07-23" author: "Fraudox Team" tags: ["Impersonation", "Threat Research"] canonical: "https://fraudox.com/blogs/deepfake-ads-using-your-founders-face" publisher: "Fraudox" publisher_url: "https://fraudox.com" license: "Free to quote with attribution to the canonical URL." --- # Your founder's face in an ad you never bought > A deepfake founder ad is removed by targeting the funnel: the advertiser account paying for delivery, the landing page and the domain, not just the video creative. *Published July 23, 2026 by Fraudox Team. Canonical version: https://fraudox.com/blogs/deepfake-ads-using-your-founders-face* A deepfake ad is a media buy, not a video file. A synthetic clip of your founder endorsing a trading platform reaches almost nobody on its own: it reaches people because an advertiser account pays for delivery. The removal target is therefore the whole funnel: the creative, the account funding it, the landing page, the domain under it, and the messaging handle where an operator closes the victim. The claim that moves fastest is rarely "this is a deepfake". It is the platform's own financial advertising policy, filed alongside a likeness complaint from the depicted executive and a trademark claim on the marks in the creative. ## The funnel, layer by layer Each report reaches one layer, which is why a single filing so often changes nothing. | Layer | What it is | What removing it reaches | | --- | --- | --- | | Creative | The video, headline and button | One ad. The account uploads the next cut. | | Advertiser account | The account funding delivery | Every creative it runs, including ones not yet uploaded | | Landing page | The fake broker or "investment platform" | The conversion step, via a [fake page takedown](https://fraudox.com/fake-pages-takedown) | | Domain | The registered name and its host | The whole site, once the [registrar](https://fraudox.com/glossary/registrar) suspends it | | Handoff channel | The WhatsApp or Telegram account working the victim | The operator, via the app's abuse process | ## Why nobody inside the company sees it An organic upload usually surfaces quickly, because your own audience tags you. Paid distribution does the opposite. The ad never touches your channels, and it is bought against an audience picked to exclude anyone likely to recognise the fraud: markets where your founder is known but your staff are not, age bands that answer investment offers. Brand-mention monitoring will not find these campaigns, because the copy may never name you in text. Three things do. Query the platform ad transparency libraries by advertiser name and by your executive's name, accepting that what each library publishes varies. Check from the geographies being targeted rather than from head office. Treat one customer asking whether the endorsement was real as a signal rather than noise. ## Three claims platforms have machinery for 1. **Unauthorised use of a likeness.** [Executive impersonation](https://fraudox.com/glossary/executive-impersonation) reports come from the depicted person or an authorised representative, and generally require identity documents and a statement that no consent was given. 2. **Trademark misuse in the creative.** Logos, wordmarks and fabricated news branding go to the brand rights process, which wants registration numbers and jurisdictions. That is [brand impersonation](https://fraudox.com/glossary/brand-impersonation), not an authenticity question. 3. **Breach of financial advertising policy.** Ad platforms restrict investment promotions in many markets to advertisers who have completed verification or hold the relevant authorisation there. An unverified advertiser running them breaches that rule whether or not the video is synthetic. > The policy breach is the fastest of the three: it asks the platform to check its own rule against its own advertiser rather than adjudicate a dispute between strangers, and it is enforced at the account level. ## Evidence that is specific to ads 1. The ad library entry and its identifier, where the platform publishes one. 2. The paying advertiser: account name, numeric ID, disclosed payer and country. 3. The creative in full: video file, thumbnail, headline, button label. 4. The landing URL, the redirect chain, and WHOIS and hosting records for the domain. 5. Targeting as disclosed: geography, language, age bands, run dates. 6. The money path: the payment page, the account or wallet named, the handoff handle. 7. A dated statement from the executive that no endorsement was given. That sits on top of [the standard evidence pack](https://fraudox.com/blogs/the-evidence-pack-what-platforms-need). Ask for the account, not only the creative: one funded account can run any number of creatives, so removing a single video leaves the operator free to upload the next variant that day. The same logic as [going after the profile rather than each post](https://fraudox.com/case-studies/executive-impersonation-social-removal). ## What detection tooling cannot carry Do not build the case on an automated authenticity score. Deepfake detectors disagree with one another, degrade on compressed re-uploads, and are contested enough that a reviewer can set them aside. What holds up is duller: the executive's own statement, an advertiser not authorised to promote investments in that market, a domain registered weeks ago, a payment path leading somewhere the real company does not bank. The video is the part everyone stares at and the least durable part of the case. Whoever built the funnel can regenerate the clip in an afternoon. What they cannot cheaply rebuild is a verified advertiser account, a warmed payment method and a domain that still resolves, so aim there: strike the account on the [social platform](https://fraudox.com/social-media-takedown), pull the landing page, take the name at the registrar. That is the difference between a removed video and a stopped campaign. --- Published by Fraudox (https://fraudox.com), a brand protection takedown service that removes phishing sites, impersonating pages, fake social accounts, counterfeit apps, scam domains and stolen content, and bills only for confirmed removals.