Help · Deepfake laws

Deepfake laws: what the law says, country by country

Lawmakers are moving fast against non-consensual deepfakes, and platforms now have clear duties to act on reports. Here is the essential, in plain English.

The main laws

WhereLawWhat it says
United StatesTAKE IT DOWN Act (2025)Knowingly publishing non-consensual intimate images, real or AI-generated, is a federal crime. Covered platforms must remove them within 48 hours of a valid request. The FTC has enforced the platform duties since May 2026.
United StatesState lawsMany states have their own laws on intimate deepfakes, election deepfakes, online impersonation and the right of publicity (for example California, Texas, New York, Tennessee).
United KingdomOnline Safety Act 2023Sharing, or threatening to share, an intimate image without consent is a criminal offence, including deepfakes. Platforms have legal duties to tackle illegal content.
European UnionDigital Services ActHosting services must provide a way to report illegal content and handle reports diligently. Large platforms must offer an internal appeal against their decisions.
European UnionAI Act, article 50Since 2 August 2026, AI-generated or manipulated deepfakes must be disclosed as such.
European UnionGDPRYou can ask for personal data processed unlawfully to be erased, and for search results about you to be delisted.
FranceCriminal Code, art. 226-8, 226-8-1, 226-4-1Non-consensual deepfakes: up to 2 years and €45,000 online. Sexual deepfakes: up to 3 years and €75,000 online. Impersonation: 1 year and €15,000.

Depending on the case, other offences may apply: fraud, harassment, blackmail, invasion of privacy or copyright infringement.

Platform rules go further

Even where the law is silent, platform policies are not. Meta, TikTok, YouTube, X, Reddit and Google all ban non-consensual intimate imagery, and most ban impersonation and deceptive synthetic media. A report that cites the right policy, with evidence, is often faster than any court.

Your options

  1. Report to the platform or host, then appeal if refused.
  2. Use the dedicated legal routes where they exist: TAKE IT DOWN Act requests in the US, DSA notices in the EU.
  3. Ask search engines to delist the pages.
  4. Report it to the police and, where appropriate, take legal action with a lawyer (including urgent court orders).

General information, not legal advice. Laws differ from country to country: for advice on your situation, talk to a lawyer where you live.

You do not have to do this alone

We find, capture, report and chase every item for you, then keep watch so it does not come back. Unlimited reports, one subscription.

See pricing

Talk to us in confidence

Tell us what is going on in a few lines. We reply with a free first assessment of what can be done.

  • Reply within 1 business day
  • Strictly confidential
  • No commitment

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

Frequently asked questions

Is a deepfake made as a joke illegal?
It depends on the country. Obvious parody or a clear AI label can change the analysis for ordinary deepfakes, but sexual deepfakes of real people without consent are illegal in a growing number of places regardless of any label.
Can I get compensation?
Often, yes: victims can usually claim damages in civil proceedings, and sometimes as part of a criminal case. A lawyer in your country can advise you.
Do platforms have a deadline to remove content?
In the US, the TAKE IT DOWN Act sets 48 hours for non-consensual intimate images. In the EU, the DSA requires diligent handling without a single deadline for all content. In practice, timing varies by platform and by how serious the content is.

Sources and references

  1. US Federal Trade Commission, “Take It Down Act enforcement starts now”
  2. Regulation (EU) 2022/2065, Digital Services Act, EUR-Lex
  3. Regulation (EU) 2024/1689, AI Act, EUR-Lex
  4. Regulation (EU) 2016/679, GDPR, EUR-Lex
  5. French Criminal Code, Légifrance (in French)