Table of Contents
Mexico AI Deepfake Law: What Families Need to Know
Mexico AI deepfake law explained: the 2026 reform on AI-made intimate images, the Ley Olimpia rules already in force, and how families report it and get help.
In February 2026, Mexico’s Chamber of Deputies voted unanimously to make one thing explicit: creating a sexual image, video or audio of someone with AI, without consent, is digital violence. The Mexico AI deepfake law reform adds that language to the federal law on women’s right to a life free of violence, and it covers sharing such material even when it was “created totally or partially” by technology. It went to the Senate. As of November 2026, it does not appear in the official consolidated text of the law, so it is not yet in force.
That does not leave families unprotected. Mexico’s 2021 “Ley Olimpia” reforms already punish sharing intimate images without consent, including fake ones attributed to a real person, and the federal penal code treats sexual material of anyone under 18 as a serious crime. Here is what the new reform changes, what already applies, and what to do if this happens to your child.
Key Takeaways
- The February 2026 reform is not law yet. It passed the Chamber of Deputies unanimously and went to the Senate; the law’s official text, last amended January 15, 2026, does not include it.
- Sharing fake intimate images is already a federal crime for adults: three to six years under Article 199 Nonies of the Federal Penal Code, part of the 2021 Ley Olimpia package.
- For minors, Article 202 applies, with seven to twelve years for producing or distributing sexual material involving under-18s.
- Prosecutors and judges can order platforms to remove content under Article 20 Sexies of the women’s violence law.
- Never download, save or forward the image, even to report it. Under Article 202, reproducing or storing that material can itself be a crime.
If your child is in crisis or talking about harming themselves: in the US, call or text 988 (Suicide & Crisis Lifeline) or text HOME to 741741 (Crisis Text Line). In Mexico, call LĂnea de la Vida, 800 911 2000 (free, 24/7). Elsewhere, find a line at findahelpline.com. In an emergency, call 911.
What the 2026 Mexico AI deepfake law reform says
The reform is an addition to Article 20 Quáter of Mexico’s General Law on Women’s Access to a Life Free of Violence. According to N+ coverage of the vote, the new paragraph would sanction any intentional “total or partial generation” of intimate sexual images, videos or audio that simulate sexual acts or pornographic depictions of a person, without consent, “through the use of technological means.” It also covers disclosing, distributing, publishing or sharing that material.
In the same session, deputies approved Federal Penal Code changes on sexual harassment and stalking: one to three years in prison, with the penalty raised by up to half when the victim is a minor. All ten measures passed unanimously and were sent to the Senate. Deputy Annia Sarahà Gómez (PAN) framed the point plainly: when generative AI is used “to fabricate false intimate images, to humiliate, to extort or destroy reputations, it stops being innovation and becomes violence.”
Mexico has no general AI law yet. A separate initiative for a General Law on Artificial Intelligence, with a chapter on children, was in committee alongside four other AI bills as of July 2026, La CrĂłnica reported.
The Ley Olimpia rules already in force
“Ley Olimpia” is the popular name for reforms named after activist Olimpia Coral Melo, whose intimate video was shared without her consent in Puebla. States adopted versions first. The federal package was published on June 1, 2021, and it is the base the 2026 reform builds on.
Reading the current Federal Penal Code, three articles matter most:
- Article 199 Octies punishes anyone who shares intimate sexual images of an adult without consent, and anyone who records, photographs or “elaborates” such images without consent: three to six years in prison and a fine.
- Article 199 Nonies applies the same penalty when the intimate images shared “do not correspond with the person who is pointed out or identified in them.” That is the closest existing fit for a deepfake of an adult.
- Article 199 Decies raises penalties by up to half in several cases, including when the offender is a current or former partner, and when the victim attempts to harm themselves as a result.
The women’s violence law already defines digital violence as sharing “real or simulated” intimate content without consent. Its Article 20 Sexies lets prosecutors and judges order platforms, by email or in writing, to interrupt, block or delete the content immediately, while preserving it as evidence.
What changes for children
When the person shown is under 18, the relevant crime is different and heavier. Article 202 punishes producing, photographing, filming or distributing sexual material involving minors, “real or simulated,” with seven to twelve years. Article 202 BIS punishes storing it.
Whether a fully AI-generated image of a real child fits that wording has not, as far as we could find, been settled by Mexico’s federal courts. Argentina faced the same question and, as Chequeado explains, one appeals court read its law to cover AI-generated material in 2025, but Mexico has no equivalent ruling we could locate. The 2026 reform would remove much of the doubt by naming generation through technology outright.
How “nudify” deepfakes are made, and why the law cares about creation
A sexual deepfake is a fabricated image, video or audio clip that shows a real person in sexual content they never took part in. Most of the images reported in schools come from “nudify” apps. Technically, these are image-generation models: the software takes an ordinary photo, masks the region covered by clothing, and has a model trained on explicit imagery predict new pixels to fill it. Nothing is “revealed.” The output is invented, which is why the law calls it “simulated.”
That distinction explains the reform’s focus. Older laws were written around sharing a real recording. A fabricated image can do the same damage without any real recording existing, and it can be made in seconds from a school photo or a social media post.
The research on how common this is among young people is still early.
- Thorn’s 2025 survey of young people aged 13 to 20 in the United States found that 1 in 17 teens reported having deepfake nudes made of them, and about 1 in 8 young people personally knew someone who had been targeted while under 18. Most young people (84%) said deepfake nudes harm the person depicted. Thorn is a child-safety nonprofit; this is a single online survey, and US-only.
- A peer-reviewed study by Umbach, Henry, Beard and Berryessa, presented at the 2024 ACM CHI conference, surveyed more than 16,000 adults in 10 countries. About 2.2% reported being victimized by deepfake pornography and 1.8% reported perpetrating it. Respondents in countries with specific laws still reported both, which the authors read as a sign that laws alone do not deter.
That second finding is worth keeping in mind. A Senate vote in Mexico would matter. It would not, by itself, stop a classmate with an app.
How reporting works today, step by step
| Step | Where | What it does | Notes |
|---|---|---|---|
| Report on the platform | The app’s report button (choose non-consensual intimate content) | Starts removal | Do this first; note the URL and username |
| Hash the image (under 18) | NCMEC’s Take It Down | Lets participating platforms find and remove copies | Works worldwide; the image never leaves the device it is on |
| Hash the image (18+) | StopNCII.org | Same idea for adults | Run by a separate nonprofit |
| File a criminal complaint | Your state prosecutor’s office (FiscalĂa) or the FGR | Opens an investigation; a prosecutor can order takedowns under Art. 20 Sexies | Bring screenshots of profiles, dates and messages, not of the image |
| Tell the school | Principal or counselor | Stops the spread among classmates | Ask what the school’s protocol is in writing |
| US families | Platform request under the TAKE IT DOWN Act; FTC complaint if ignored | Platforms must remove within 48 hours of a valid request | US law only; enforced since May 19, 2026 |
Take It Down creates a “hash,” a digital fingerprint, on your device; only that fingerprint goes to NCMEC, and participating platforms scan their public or unencrypted services for matches. It does not work on encrypted chats such as WhatsApp messages. NCMEC is explicit: do not download or share an image in order to submit it.
For US readers, our guide to using the Take It Down Act to get a deepfake of your child removed covers the federal process in detail.
What to do at home
Have the conversation before anything happens
Tell your child, in a calm moment, that fake sexual images of real people exist, that making or forwarding one hurts a real person and can be a crime, and that if it ever happens to them or a friend, they can tell you without being punished. Thorn’s survey found that more than half of young victims sought offline support. The first person they tell shapes what happens next.
Make “never forward” a family rule
Most harm comes from spread. A child who receives an image should not forward it, even to “warn” someone or show an adult. They should tell an adult and report it on the platform.
If it happens: preserve evidence without the image
Write down or screenshot the usernames, profile links, group names, dates and any messages, but not the image itself. Then report on the platform, tell the school, and go to the prosecutor’s office. Our step-by-step guide on what to do in the first 48 hours if your child is a victim of a sexual deepfake walks through each step.
Watch the emotional side
Being the subject of a fake sexual image can bring shame, fear and withdrawal, even when everyone knows it is fake. Keep your child in school routines if you can, ask the school counselor for support, and talk to their doctor if sleep, appetite or mood change.
What not to do: open, save or send the image
Do not ask to see it, download it, screenshot it or forward it, not even to the police or the school. Under Article 202, reproducing or storing sexual material of a minor can be a crime regardless of intent, and every copy is one more to remove. Let investigators collect it.
What to Watch For Over the Next 3 Months
- Week 4: Whether the Senate schedules the Article 20 Quáter reform. Publication in the Diario Oficial de la Federación is what makes it law.
- Month 2 red flags: Your child suddenly avoiding school or a group chat, deleting accounts, or anxious when their phone buzzes. These can be signs of image-based abuse or sextortion; our sextortion guide for parents covers the overlap.
- Month 3 self-check: Does your child know they can tell you, and that you will not take their phone away as punishment for being a victim?
Frequently Asked Questions
Is it illegal in Mexico to make an AI nude of someone?
Sharing one is already a federal crime: Article 199 Nonies punishes sharing intimate images that do not correspond to the person shown, with three to six years. The February 2026 reform would name AI generation explicitly, but it still needs the Senate. When the person shown is a minor, Article 202’s penalties of seven to twelve years apply.
What if the person who made it is also a minor?
Mexico has a separate justice system for adolescents, with different procedures and measures. Report it anyway, to the platform, the school and the prosecutor’s office. The goal is to stop the spread and protect your child, and authorities decide how to handle a minor offender.
Can I get the image taken down without going to court?
Often, yes. Start with the platform’s report tool and, for anyone under 18, NCMEC’s Take It Down, which works worldwide. A prosecutor can also order platforms to remove content quickly under Article 20 Sexies once you file a complaint.
Should I look at the image to confirm it is fake?
No. You do not need to see it to report it, and saving or sending it can be a crime under Article 202. Your child’s word, plus the usernames, links and dates, is enough to start. Investigators will collect the image.
Where can my child get emotional support?
Start with people they trust: you, a school counselor, their doctor. If your child talks about hurting themselves, in Mexico call LĂnea de la Vida at 800 911 2000. In the US, call or text 988 or text HOME to 741741. Elsewhere, findahelpline.com lists local lines. In an emergency, call 911.
About the author
Ricky Flores is the founder of HiWave Makers and an electrical engineer with 15+ years of experience building consumer technology at Apple, Samsung, and Texas Instruments. He writes about how kids learn to build, think, and create in a tech-saturated world. Read more at hiwavemakers.com.
Sources
- N+ (Nmas). (2026, February 18). “Diputados aprueban reformas para sancionar la violencia a través de IA generativa.” https://www.nmas.com.mx/nacional/diputados-aprueban-reformas-para-sancionar-la-violencia-a-traves-de-ia-generativa/
- Cámara de Diputados. (2026). “Código Penal Federal” (texto vigente, última reforma DOF 13-03-2026), Arts. 199 Octies–Decies, 202, 202 BIS. https://www.diputados.gob.mx/LeyesBiblio/pdf/CPF.pdf
- Cámara de Diputados. (2026). “Ley General de Acceso de las Mujeres a una Vida Libre de Violencias” (texto vigente, última reforma DOF 15-01-2026), Arts. 20 Quáter and 20 Sexies. https://www.diputados.gob.mx/LeyesBiblio/pdf/LGAMVLV.pdf
- La Crónica de Hoy. (2026, July 22). “México debe regular la IA para proteger a las infancias, afirma Morena en la Cámara de Diputados.” https://www.cronica.com.mx/nacional/2026/07/22/mexico-debe-regular-la-ia-para-proteger-a-las-infancias-afirma-morena-en-la-camara-de-diputados/
- Thorn. (2025, March). “Deepfake Nudes & Young People.” https://www.thorn.org/research/library/deepfake-nudes-and-young-people/
- Umbach, R., Henry, N., Beard, G., & Berryessa, C. (2024). “Non-Consensual Synthetic Intimate Imagery: Prevalence, Attitudes, and Knowledge in 10 Countries.” Proceedings of the CHI Conference on Human Factors in Computing Systems. https://doi.org/10.1145/3613904.3642382
- National Center for Missing & Exploited Children. (2026). “Take It Down” and FAQ. https://takeitdown.ncmec.org/
- Federal Trade Commission. (2026, May 19). “FTC Begins Enforcing the TAKE IT DOWN Act.” https://www.ftc.gov/news-events/news/press-releases/2026/05/ftc-begins-enforcing-take-it-down-act
- Chequeado. (2026, October 8). “QuĂ© hacer si tu hija o hijo es vĂctima de un deepfake sexual creado con IA.” https://chequeado.com/el-explicador/que-hacer-si-tu-hija-o-hijo-es-victima-de-un-deepfake-sexual-creado-con-ia/
