What can this look like?
Fictional example: Marta, 29, is a Portuguese nurse living in Vienna. One evening an unknown account sends her a short video on a social network: her face, taken from holiday photos, on someone else's body. Unless she pays 500 euros, the message says, her colleagues will get the video.
Marta does not reply. She screenshots the profile, video and message, copies the profile link and notes the time. She reports the account and goes to the police the next morning. An advice centre then helps her with the removal requests.
What should you do in the first hour, day and week?
Order matters: if you ask for removal first, you may have no proof left for the police.
| When | What you do | What to watch |
|---|---|---|
| First hour | Save links and screenshots, list every place it appears | Do not forward, comment or argue |
| First day | Report it on each platform, send the Google request, check StopNCII | Keep confirmations and case numbers |
| First week | Get advice, consider a police report, check replies | Appeal if a report is rejected |
| Afterwards | Search for your name again every few weeks | Add any new places to the same list |
Tip: If you feel threatened or are being blackmailed, call 112. It works free of charge everywhere in the EU and is also used in Switzerland. In Germany you can also reach the police on 110.
How do you save evidence without making things worse?
You need evidence for the platform, for Google and for a possible police report. Keep it short and complete:
- the full address (URL) of each post, profile or channel
- screenshots showing the content, the profile name, the date and the time
- messages in which someone tells you about the material or threatens you
- a short note of when and from whom you heard about it
Swiss Crime Prevention advises against passing deepfakes on, even to warn others. Ask friends for the link, not the file. To keep track of everything, see how to document a digital incident.
Where do you report it so it disappears?
On the platform
Use the report function next to the post or profile. Under the EU Digital Services Act (DSA), platforms must explain their moderation decisions and let you ask for a review. If you suspect a platform is breaching the DSA, you can complain to the Digital Services Coordinator in your country.
On Google
Google removes fake content that shows you nude or in a sexually explicit situation from its search results if you can be identified and it was shared without your consent. It stays on the website itself, so report it there as well.
Google Search: request removal of fake sexual content
- Open the Google Search Help article on removing personal sexual content from Google Search.
- Select “Start removal request”.
- Choose “Remove fake sexual or nude content with you in it”.
- Add the URLs of the pages and upload screenshots. You can edit them so that only your face is visible.
- Submit the request. If a result is removed, Google lets you know by email.
Menu path as given in the manufacturer’s help; it may differ by device and version: support.google.com
With StopNCII
StopNCII.org creates a digital fingerprint (hash) of an intimate image on your own device, including AI-generated images. Only the hash is shared with participating platforms so they can detect copies; the image never leaves your device. You must be the person in the image, have been 18 or over in it and have the file yourself. It only works on public-facing platforms, not in encrypted messaging services. If you were under 18 in the image, StopNCII refers you to Take It Down, even if you are an adult now.
What applies in each country?
Whether a deepfake is a criminal offence depends on the country and the content. The box shows where to turn (as of 18 October 2026).
| Country | Where to turn | Legal route or note | Emergency |
|---|---|---|---|
| Germany | Police; HateAid (free counselling) | Depending on the case, e.g. insult or defamation; a specific offence for sexualised deepfakes is only a government draft (adopted by the Federal Cabinet on 7 October 2026) | 110 or 112 |
| Austria | Police; district court for an injunction; report on the platform | Fast-track injunction (Mandatsverfahren) at the district court without a prior hearing if a breach of your personality rights violates your human dignity; no lawyer required, court fee 148 euros (as of 1 August 2026) | 112 |
| Switzerland | Cantonal police | Swiss Crime Prevention names e.g. Art. 197a and 179decies Criminal Code and personality rights (Art. 28 Civil Code) | 112 |
| France | Pharos for serious illegal content; police or gendarmerie; CNIL for misuse of your personal data | Art. 226-8 Criminal Code: montage of a person's image without consent, up to one year in prison and a 15,000 euro fine, more in some cases | 112 |
| Italy | Polizia Postale | Art. 612-quater Criminal Code (Law 132/2025): one to five years in prison, usually on the victim's complaint | 112 |
| Spain | AEPD priority channel; INCIBE helpline 017 | AEPD: your face is personal data even if the rest is not real | 112 |
Switzerland is not an EU member, so the DSA and the EU AI Act do not apply there automatically. Platform reporting, the Google request and StopNCII are still open to you.
This overview is not legal advice. Discuss a police report or court case with an advice centre or a lawyer.
What changes with the EU ban from 2 December 2026?
From 2 December 2026, the EU bans AI systems that generate realistic images, video or audio showing identifiable people nude or in sexual acts without their consent. The ban was added to the AI Act by Regulation (EU) 2026/1744 of 8 July 2026. It governs which tools may be offered and used. An image already online does not disappear because of it.
The AI Act's transparency rules have applied since 2 August 2026, with a transition period until 2 December 2026 for certain providers whose systems were already on the market to mark AI-generated content. What labels and watermarks can and cannot tell you is explained in understanding AI labelling.
What if someone threatens you or demands money?
Sometimes a deepfake comes with a demand: money, real photos or a meeting, or else the material will be spread. Do not engage and send nothing. Save the messages with sender and time and report it to the police. Block the person only once everything is saved.
Look after yourself too and talk to someone you trust. Later, you can review which photos of you are public: your profile through a stranger's eyes.
Common mistakes
- Forwarding the image to friends to warn them. This spreads it further.
- Asking for removal first and saving evidence later. Once the post is gone, so is your proof.
- Saving only the image, without the address, profile name and date.
- Assuming the Google request deletes the image from the internet.
- Giving in to demands or confronting the person in public.
Your checklist
The checkboxes are for your own reference. No entries are sent to us. To keep a copy, print the page or save it as a PDF using your browser menu.
Common questions
Is making a deepfake of me a crime?
It depends on the country and the content. Italy has a specific offence for spreading AI-faked images, videos or voices without consent if this causes unjust harm; in Germany such an offence for sexualised deepfakes is still a draft. You can report it to the police in every country. For your own case, an advice centre or a lawyer can help.
Does Google delete the image from the internet?
No. Google can only remove results from its own search pages. The page itself remains available through a direct link or other search engines, so also report it to the platform or website owner. The request is still worth it: people searching for your name are less likely to come across the material.
Can I use StopNCII if I do not have the image?
No. StopNCII needs the file on your device to create the hash, and you must be the person in the image. You cannot submit a hash on someone else's behalf. Without the file, you still have the platform report, the Google request with links and screenshots, and a police report.
What can I do if the platform does not remove it?
In the EU, you can ask for a review through the platform's internal complaint system or turn to an out-of-court dispute settlement body. If the platform breaches the DSA, complain to your country's Digital Services Coordinator. In Switzerland, where the DSA does not apply, a police report and civil action remain.
Sources and further information
Primary sources from public authorities, consumer organisations and providers. A source may be in another language; follow the guidance that applies to your service and where you live.
- AI Act Service Desk (European Commission): Timeline for the implementation of the EU AI ActSource review date:
- European Commission: Your rights under the Digital Services ActSource review date:
- StopNCII.org: Frequently asked questionsSource review date:
- European Commission: 112, the European emergency numberSource review date:
- Austrian Federal Ministry of Justice: Applying for an injunction in fast-track proceedings (German)Source review date:
- CNIL: Hypertrucage (deepfake), protecting yourself and reporting (French)Source review date:
- Gazzetta Ufficiale No 223 of 25 September 2025: Law No 132 of 23 September 2025 (PDF, Italian)Source review date:
- AEPD: Step by step, requesting removal of sexual or violent content via the Priority Channel (Spanish)Source review date:
