How to Report and Remove a Deepfake of Your Executive
A deepfake of your CEO is running as an ad right now. Here's exactly what to do: the reporting steps, the evidence platforms need, and how to stop it from coming back.
To remove a deepfake of your executive: capture evidence first, report through each platform’s dedicated impersonation and fraud channels (not the generic “report post” button), frame it in policy language, and track every variant to confirmed removal while watching for the rebound. Speed is everything. Every hour the impersonation runs, it costs more victims and more brand damage.
If a deepfake impersonating your executive is live, you are on the clock. Every hour it runs, it reaches more victims and does more damage to your brand. This guide walks through exactly how to get it removed and, just as important, how to keep it from reappearing.
Step 1: Capture evidence before you do anything
The instinct is to report first. Don’t. Capture first. Once an account is taken down, the evidence can disappear with it.
- Screenshot the content, the account or page running it, and any URLs.
- If it’s a paid ad, note where it’s running and save the ad’s link or library entry.
- Record the date, time, and reach if visible.
This evidence is what turns a slow, generic report into one a platform’s review team can act on quickly.
Step 2: Report through the right channel
Generic “report post” buttons are the slowest path. Platforms maintain dedicated channels for impersonation and fraud that move faster:
- Meta (Facebook & Instagram): Use the impersonation and “fraudulent ad” reporting flows, and check the Ad Library to find every variant of the campaign, not just the one you saw.
- YouTube & TikTok: Use the impersonation and misleading-content reporting tools, which are distinct from generic spam reports.
- X and others: File under impersonation and platform-manipulation policies.
The key insight: most campaigns run many variants at once. Reporting one ad rarely stops the campaign; you have to find and report the whole cluster. (If the impersonation is running as paid ads, see how deepfake ad fraud is structured.)
Step 3: Escalate with the right framing
A report that says “this is fake” gets triaged slowly. A report that documents policy-violating impersonation of a real, named individual, with evidence attached, gets actioned faster. If you’re not yet certain the content is synthetic, start with how to spot a deepfake. Frame it in the platform’s own policy language: impersonation, fraud, scam, synthetic media.
Step 4: Track to confirmed removal, and watch for the rebound
This is the step most teams miss. Removal isn’t the end:
- Confirm the content is actually gone, not just hidden from you.
- Watch for the same campaign relaunching under new accounts and new creatives within days. Organized networks expect takedowns and plan around them; it’s why executive impersonation thrives on social platforms.
- Re-file as new variants appear.
A one-time takedown is not protection. Continuous removal is.
The bottom line
Removing a deepfake is a process, not a single click: capture evidence, report through dedicated channels, frame it in policy terms, and track it to confirmed removal while watching for the rebound.
Doing this manually, around the clock, across every platform, doesn’t scale against automated campaigns. That’s the problem Revelum was built to solve, detecting impersonations at scale and filing tracked takedowns in under 24 hours, then monitoring for re-uploads.
We will assess your situation and tell you what we are seeing, typically within 24 hours.
Frequently asked questions
- How long does it take to remove a deepfake?
- It varies by platform and by how well the report is documented, from hours to weeks. Well-evidenced reports filed through dedicated impersonation channels move fastest. Revelum processes takedowns in under 24 hours.
- What if the platform rejects my report?
- Re-file with stronger evidence and explicit policy framing, and escalate through business or verified-rights channels where available. Persistence and documentation matter.
- Can I get ahead of this instead of reacting?
- Yes. Continuous monitoring detects impersonations as they launch, often before they gain reach, so takedowns start early instead of after the damage is done.