Deepfakes

Get a deepfake removed

To remove a deepfake, you report it to every site that hosts it, ask Google to drop it from search, follow up and watch for copies. We do all of it for you, without you having to look at the content again, then check every day that it does not come back.

  • A specialist checks every case
  • Daily monitoring, same-day reports
  • Fixed prices, shown upfront
Black and white face of a woman, the right half breaking into squares

What we can do for you

Two ways to hand us your case. You choose, we take care of the rest.

When every hour counts

Emergency Response

€290 one-off

  • Open a case, get a dedicated handler within the hour, 8 am to 4 pm Eastern Time
  • First reports within 2 hours
  • Follow-up every day for 7 days
Open my case

One-off payment, on the next page.

Monthly: we report, then we keep watch

Personal Shield

€29 a month

or €199 a year

  • Up to 5 items reported a month, the same day
  • Daily monitoring of your name, your photos checked at every check-up
  • Full check-up at sign-up, then every month, with a report
Start the Personal Shield

Renews every month, stop online at the end of the current month. Ages 18 and over. Terms

Not sure? Tell us what is happening: our team replies within 24 hours, free of charge. Free assessment

A professional or a business? Pro Shield, from €99 · Business Shield

The steps to remove a deepfake

  1. Keep the evidence. The page address and a dated screenshot, before reporting and without sharing the content.
  2. Report on the right grounds. Every platform that hosts it gets a request on the most specific grounds: intimate content without consent, AI-generated content or impersonation.
  3. Get it off Google and Bing. Alongside the report to the site, a delisting request aims to stop the page from showing up when someone searches your name.
  4. Follow up. With no answer, we follow up and use the appeal routes set out in the EU Digital Services Act (DSA).
  5. Remove the copies. Every day we check your name and the links already handled. Every copy we find is reported the same day.

You can do these steps yourself with our guide deepfake: what to do. If you would rather hand it over, we do it for you.

What is a deepfake?

A deepfake (also spelled "deep fake") is an image, video or audio recording generated or edited by an AI model so that a real person appears to say or do something they never said or did. A few years ago this took technical skill. Today, consumer apps do it in seconds from a single photo or a few seconds of voice.

It is growing fast: the UK Home Office estimates that 8 million deepfakes were shared in 2025, up from 500,000 in 2023.

The main types of deepfakes

  • Face swaps: your face placed on someone else’s body in a photo or video.
  • Intimate deepfakes (“deepnudes”): fake nude or sexual images made from ordinary photos. One of the most violent forms. Read our guide.
  • Voice clones: your voice imitated in a message, a video or a phone call. Read our guide.
  • Scam ads: your face, sometimes animated, promoting a fake investment or product. Read our guide.
  • Altered videos: real footage with the words or gestures changed.

Why a deepfake needs fast action

Your reputation

Once shared and indexed by Google, a fake can follow every search of your name.

Your dignity

An intimate deepfake is abuse, even when everyone knows it is fake.

Blackmail and scams

Fakes are used to pressure you or to deceive your family and audience.

Your wellbeing

Stress, shame, sleepless nights: not having to face forms and refusals alone matters.

Trust

Clients, employers, followers: the further it spreads, the harder the truth is to restore.

Copies

Every hour online multiplies screenshots and reuploads on other sites.

Free first assessment

Is this happening to you right now?

Tell us what is going on in a few lines.
We will tell you what can be done.

  • Reply within 24 hours
  • Read by our team, in confidence
  • Free, no obligation to buy

Emergency? With Emergency Response, a dedicated handler within the hour, 8 am to 4 pm Eastern Time. See Emergency Response

Never send us an intimate image. A link or a description is enough.

We will never ask for your password, or for a payment in cryptocurrency or outside our Stripe payment page, and we never negotiate with a blackmailer. Our emails only come from contact@sosdeepfake.com. How to spot a fake removal service

You are
Where do you stand?

Emergency Response €290 one-time payment

Handled within the hour. We launch the searches, report, follow up relentlessly and keep watch so nothing comes back.

Open my case

Every day, 8 am to 4 pm Eastern Time.

Or tell us first, for free: a written reply within 24 hours.

Who are we looking for?

Who are you?

Your city is only used so we do not mix you up with someone who has the same name.

Your company

To spot a company set up under a name close to yours.

In the US, your EIN. In the UK, your Companies House number. In France, your SIREN or SIRET.

We check addresses that imitate yours and how well your emails are protected against spoofing.

Which accounts are really yours?

We need them to recognize your real accounts and spot the ones that imitate them.

We check addresses close to yours.

They are the ones scammers imitate in CEO fraud.

What brings you here

What is happening?

We look at it first. No link? Choose “I want to act before anything happens”, and we will search for you.

The link to the fake account, page or video, if you have it. Otherwise, tell us where you saw it.

Where should we send the result?

Where can we reply?

The result is sent to this address, and only to this one.

An address on your website’s domain. A Gmail or Yahoo address does not let us check that you represent the company.

One free check-up per person and per company. Never a photo or an ID document. Your answers are deleted within 30 days (privacy).

How we act against your deepfakes

Reverse image search

With your consent, we run reverse image searches (Google Lens, Bing) to find copies of your photos.

Human review

A person confirms every item before it is reported. No abusive reports, better removal rates.

Dated evidence

Each page is captured with its date before it is reported, so the proof survives removal.

Reports on the right grounds

Non-consensual intimate imagery, AI-generated content, impersonation, personal data: each request cites the rule that applies (DSA, TAKE IT DOWN Act, GDPR, platform policy).

Search delisting

Requests to Google and Bing so the content no longer shows up when people search your name.

Daily monitoring

Every day, automated searches check your name and the links already handled. Every week (every month with the Personal Shield), a full check-up of your online presence. Every copy we find is handled in turn.

Where we file reports

Instagram, Facebook, TikTok, YouTube, X, Snapchat, LinkedIn, Telegram, Discord, Reddit, OnlyFans, MYM, Google, marketplaces, forums and websites. Found it somewhere else? Send us the link: we will find who to report it to.

What the law says

WhereRuleWhat it means for you
United StatesTAKE IT DOWN Act (2025)Publishing non-consensual intimate images, including AI-generated ones, is a federal crime. Covered platforms must remove them within 48 hours of a valid request. The FTC has enforced this since May 19, 2026.
European UnionDigital Services ActHosting services must run a notice-and-action system and handle reports of illegal content diligently. Online platforms, except micro and small companies, must offer an internal appeal.
European UnionAI Act, article 50Since August 2, 2026, deepfakes must be disclosed as AI-generated or manipulated.
United KingdomOnline Safety Act 2023Sharing, or threatening to share, an intimate image without consent is a criminal offense, including when the image is a deepfake.
FranceCriminal Code, art. 226-8 and 226-8-1Sharing AI-generated content of a person without consent is an offense. For sexual deepfakes, up to 3 years in prison and a €75,000 fine when shared online.

Many other countries and US states have their own deepfake, harassment or privacy laws. Platform rules usually go further than the law: most major networks ban non-consensual intimate imagery and deceptive synthetic media outright. Read our guide to deepfake laws.

General information, not legal advice. For advice on your situation, talk to a lawyer in your country: we can refer you.

Who it is for

Creators and influencers

Your face is your working tool, and also the most frequent target. See our page for public-facing professionals.

Public figures

Journalists, politicians, athletes: visibility attracts fakes.

Artists and actors

Your image and voice deserve the same protection as your work.

Executives and companies

Fake statements, fake announcements, CEO fraud: see our brand protection.

Agencies

Protect every talent you represent, with a case log per person.

Anyone targeted

You do not need to be famous: a few published photos are enough to make an intimate deepfake.

Frequently asked questions

Can a deepfake be removed for good?
Yes, it can be taken down from every site that hosts it and from Google results. The hard part is copies: the same content often gets uploaded again somewhere else. That is why we do not stop at the first takedown: we search every day and report every new copy.
How do I get a deepfake of me removed?
Report it to every platform that hosts it using the most specific reason, ask Google to remove it from search, follow up if there is no answer, then monitor for copies. We do all of it for you, and you never have to look at the content again.
How are deepfakes made?
With generative AI models that learn the features of a face or voice from photos, videos or recordings, often taken from social media.
Are deepfakes illegal?
It depends on the country and the content. Non-consensual sexual deepfakes are illegal in the US (TAKE IT DOWN Act), the UK, France and a growing list of other countries, and banned by the policies of all major platforms.
See 5 more questions
How do you find deepfakes?
By searching for your name and profiles, by reverse image search (with your consent), and through the links you send us. A person checks every result before anything is reported.
How long does removal take?
We file the first reports the same day. The next day for cases received after 2 pm, or after 10 pm with the Enhanced Shield (Eastern Time). Response time then depends on the platform: under the TAKE IT DOWN Act, covered US platforms have 48 hours for intimate images, while smaller sites often need more follow-ups. The final decision belongs to the platform.
How much does deepfake removal cost?
From €29 a month for an individual (Personal Shield, up to 5 items reported a month, or €199 a year). Unlimited reports with the Pro Shield: €99 a month, €499 for 6 months or €899 for 12 months. Compare plans.
Can I report it to the police?
Yes, and in many countries you should. Our dated captures can be attached to your report.
Which platforms do you cover?
The main social networks, forums, video sites, messaging apps such as Telegram, and search engines: we file reports there. Our searches run on Google, Bing and social networks. For a channel or a group, send us the link.

In short

In short, 6 questions
Who does the work?
SOS Deepfake files the reports for you: you fill in a form, a person checks it, then reports the content to the platform, the host or Google on the right grounds, and follows up until there is an answer.
How fast?
Open a case, get a dedicated handler within the hour, 8 am to 4 pm Eastern Time. With Emergency Response, the first reports go out within 2 hours, then the follow-up lasts 7 days. The takedown decision belongs to the platform or the host: SOS Deepfake commits to the work, dated evidence, the right grounds, follow-ups and appeals until there is an answer.
Is it confidential?
Yes. We never ask for an ID document or an intimate image. Your file is deleted when the service ends. We keep only the invoices, as the law requires, and what is needed to resume it for 3 months (privacy).
What if the platform says no?
We challenge the decision through the free internal complaint provided by the EU Digital Services Act (DSA, Article 20), then before a certified dispute settlement body (Article 21), or we go through the host. For a criminal complaint, we point you to the police or an independent lawyer.
How much does it cost?
Fixed, public prices: Personal Shield from €29 a month, Emergency Response €290 as a one-off payment, no subscription (all prices). The guides and the assessment of your situation are free.
Who is it for?
Individuals, professionals (doctors, lawyers, creators, elected officials) and businesses, wherever you live: platforms and Google accept reports from every country.

Act against deepfakes today

Every day online multiplies the copies. Start your protection: anything we find is reported the same day.

Get this deepfake removed

Sources

  1. UK Home Office, “Government leads global fight against deepfake threats”, February 5, 2026
  2. US Federal Trade Commission, “Take It Down Act enforcement starts now”
  3. Regulation (EU) 2022/2065, Digital Services Act, EUR-Lex
  4. Regulation (EU) 2024/1689, AI Act, EUR-Lex
  5. Regulation (EU) 2016/679, GDPR, EUR-Lex
  6. Google, “Remove personal sexual content from Google Search”