AI-generated images or deepfake videos of you on Instagram or Facebook: the removal routes

Meta does not have a single “deepfake” rule. Whether an AI-generated image or video of you comes down depends on what it shows and how it is used. Sexual or nude deepfakes are removed as non-consensual intimate imagery, a policy that explicitly includes AI-generated images. Fakes made to mock or degrade a private person fall under the bullying rules, fakes used to pose as you fall under impersonation, and a fake that breaks none of these may only get a label.

This guide maps each kind of deepfake to the policy that removes it, the report to use, and the legal layer behind it. If you are being threatened or blackmailed with fake images, contact the police first.

The short answer

• Sexual or nude deepfakes: report as nudity or sexual activity involving you without consent. Meta’s policy covers AI-generated imagery. Add StopNCII.org to block re-uploads.
• Deepfakes made to humiliate you: report as bullying or harassment, yourself. Private adults are protected from unwanted manipulated imagery when they self-report.
• Deepfakes used to pose as you, or in scam ads: report the account for impersonation, and the ads as scams.
• Political or “funny” fakes that break no other rule: Meta may label them rather than remove them.
• In the US: covered platforms must remove nonconsensual intimate deepfakes within 48 hours of a valid request under the TAKE IT DOWN Act.

Route 1: sexual or nude deepfakes

Meta’s adult sexual exploitation policy prohibits sharing, threatening to share or offering non-consensual intimate imagery where the person is “(near) nude, engaged in sexual activity or in a sexually suggestive pose”, and adds in brackets: “this includes digitally created or AI-generated imagery.” One of the listed signals that consent is missing is a “report from a person depicted in the image”. The same policy prohibits promoting or offering “nudify” apps and services that make such images.

So you do not need to prove the image is fake. Real or generated, an intimate image of you shared without consent falls under the same rule.

How to act:

1. Report every post and the account. Use the report option, choose Nudity or sexual activity, and follow the on-screen options that describe an image of you shared without consent. Our guide to Meta’s priority channel for intimate images walks through the reports and the dedicated form.
2. Hash the image. StopNCII.org says that if a deepfake or synthetic image is of you, you have access to it and it is nude or semi nude, you can hash it. The image stays on your device; only a digital fingerprint is shared with participating platforms. The service is for adults; if you were under 18 in the image, it points you to Take It Down, the service for images of people under 18.
3. Expect re-upload matching. Meta’s help page on intimate images says that after removal it uses photo-matching technology to help stop future attempts to share the image on Instagram, Facebook and Messenger.

Meta’s bullying policy adds a second basis: it removes “derogatory sexualized photoshop or drawings” of anyone, public figure or not.

Route 2: deepfakes made to mock or humiliate you

Not every fake is sexual. Face swaps into degrading scenes, edited videos that make someone look drunk or foolish, and images that exaggerate a physical feature are all common forms.

Meta’s bullying and harassment policy covers much of this, with protection that depends on who you are:

• Private adults (and minors) are protected from content sexualizing another adult, dehumanizing comparisons in written or visual form, and content “manipulated to highlight, circle, or otherwise negatively draw attention to specific physical characteristics.”
• When they report it themselves, private adults are also protected from “unwanted manipulated imagery”. This is the clause that fits most non-sexual deepfakes of private people, and it depends on a report from you.
• Public figures get less protection overall, but Meta says it removes content sexualizing a public figure once the person, or an authorized representative, confirms it is unwanted.

If a whole account exists to target you with this kind of content, our guide to reporting an Instagram hate page covers the account-level report and the tools that limit contact meanwhile.

Route 3: deepfakes used to pose as you

AI can make fake profiles more convincing: generated photos of you in new settings, cloned voices in videos, and fake endorsements. Meta’s authentic identity policy restricts or disables accounts that impersonate a person by “using their image(s), name, or likeness with the aim to deceive others”, or by speaking in their voice without authorization.

Report the account from its profile, choosing that it is pretending to be someone else. If your AI-generated face or voice appears in ads selling products or investments, the ads and the Pages behind them need reporting too; our guide to scam ads using your face or brand covers that layered approach.

What Meta usually labels instead of removing

Here is the honest limit. Meta’s misinformation policy says that for content that does not otherwise violate the Community Standards, it may place an informative label when a photorealistic image or video, or realistic audio, “was digitally created or altered and creates a particularly high risk of materially deceiving the public on a matter of public importance.”

Two consequences follow. A manipulated clip that breaks no other rule is more likely to be labeled than removed. And a fake that is merely unflattering, not sexual, not harassing under the tiers above and not impersonating anyone, may stay up. In those cases the realistic routes are legal ones.

If you live outside the US, one more route may apply: Meta’s privacy violations policy says it may remove a reported photo or video of an adult, reported from outside the United States, where applicable law may provide rights to removal. Whether your local law treats a synthetic image of you that way is a question for a lawyer.

The legal layer

In the US, the TAKE IT DOWN Act, signed on May 19, 2025, covers “digital forgery”: an intimate image of an identifiable person created or altered with software, machine learning or AI that a reasonable person would find indistinguishable from a real one. Covered platforms had one year to set up a notice and removal process. Once they receive a valid request, they must remove the image as soon as possible and no later than 48 hours, and make reasonable efforts to remove known identical copies. A valid request includes your signature, enough information to locate the image, a brief good faith statement that it is not consensual, and your contact details. The Act also makes knowingly publishing such images a crime.

Outside the US, many countries have laws on intimate image abuse, harassment and personality rights. A police report also creates a record that supports platform reports.

Evidence, and what not to do

Save the post URLs, the account names and dated screenshots before you report. Then:

• Do not reshare the fake, even to warn people. Describe it instead.
• Do not engage the poster, especially if there is a demand for money.
• Do not pay for internet-wide deletion. Official routes are free, and no outside service can remove content from Meta’s platforms on its own authority.

When to get help

A single deepfake post can usually be reported by you, today, using the routes above. Help is worth it when the fakes spread across accounts and platforms, when they are combined with impersonation or scam ads, when a business or public figure is targeted at scale, or when you need an organized evidence file for a lawyer or the police. Our unauthorized content removal service handles that work through Meta’s official reporting channels only, confidentially and with a record of each step. We make no guarantees, and we will tell you honestly if a fake falls outside what the policies remove. You can start with a free case review.