Online safety research
AI Undress Apps and Undress AI Tools: What Real Cases and Research Show
Every month, millions of people type queries like ai undress, undress app or ai undresser into search engines. What sits behind those searches is not a neutral photo trick. It is a market built on fabricating sexual images of real people without consent, and the damage is documented in court filings, police reports and peer reviewed research. This article collects the real cases and the numbers.
Key takeaways
- check_circleAn undress app uses generative AI to fabricate a nude image of a real person from an ordinary clothed photo, without that person's consent.
- check_circleGraphika counted roughly 24 million visits to 34 AI undressing services in a single month of 2023, and spam links promoting them grew by more than 2,000 percent that year.
- check_circleDocumented cases include schoolgirls in Almendralejo (Spain), students in Westfield (New Jersey) and Melbourne (Australia), and the Taylor Swift deepfake wave on X in January 2024.
- check_circlePeople who use these apps risk payment fraud, photo theft and prosecution. Victims can limit the spread with StopNCII.org and Take It Down from NCMEC.
What is an AI undress app?
An AI undress app is a service that takes a normal clothed photo of a real person and generates a fake nude or sexualized version of it. The result is synthetic. The model invents a body that was never photographed, but it keeps the person's real face, which is exactly what makes the output abusive.
These services are marketed under many names. The typical input is an ordinary photo taken from Instagram, TikTok or a school yearbook page. The typical victim never agreed to anything. A close relative is face swap porn, where a real face is placed onto existing explicit photos or videos instead of generating a body from scratch. Both belong to what researchers and prosecutors call synthetic non consensual intimate imagery, often shortened to synthetic NCII or image based sexual abuse.
How big is the AI undressing problem? Research and numbers
The problem is industrial in scale. Independent researchers tracking these services count tens of millions of visits per month, and nearly all deepfake content online is sexual and targets women.
| Source | Year | Finding |
|---|---|---|
| Graphika, report on synthetic NCII | 2023 | Identified 34 AI undressing providers with about 24 million visits in September 2023 alone. Spam links promoting these services on X and Reddit grew by more than 2,000 percent from the start of that year. |
| Security Hero, State of Deepfakes | 2023 | About 98 percent of all deepfake videos online are pornographic, and roughly 99 percent of the individuals targeted are women. |
| Sensity, deepfake census | 2019 | Found that 96 percent of deepfakes circulating online were non consensual pornography, a pattern later research confirmed at even higher levels. |
| San Francisco City Attorney, civil lawsuit against 16 undress websites | 2024 | Court filing states the 16 sites received more than 200 million visits in the first six months of 2024 and describes them as operated anonymously with no accountability to victims. |
| Internet Watch Foundation | 2024 | Confirmed a growing share of reports involving AI generated child sexual abuse imagery, including material produced from ordinary photos of real children. |
Real cases that show the damage
Statistics describe scale. Cases describe what it looks like when it happens to a specific person, classroom or school.
Taylor Swift and the deepfake wave on X
In late January 2024, AI generated explicit images of Taylor Swift spread across X. One post was viewed more than 45 million times before removal. The platform reacted by temporarily blocking searches for her name, an almost unprecedented step. The episode pushed the White House and members of Congress to comment publicly on synthetic NCII within days.
Schoolgirls as young as 11 targeted in their hometown
In September 2023, mothers in the small town of Almendralejo discovered that fake nude images of their daughters were circulating in local chat groups. More than 20 girls were affected, some as young as 11. The images had been produced from ordinary social media photos using an undress app, and some victims were blackmailed. Police identified juvenile suspects, and the case became a national scandal in Spain.
Westfield High School
In October 2023, students at Westfield High School learned that boys had used an AI tool to create fake nude images of female classmates and shared them in a group chat. Families said the school had no policy for a crime that barely existed a year earlier. The case triggered student walkouts and became one of the examples cited in the US debate over deepfake legislation.
Bacchus Marsh Grammar
In June 2024, around 50 girls at Bacchus Marsh Grammar in Victoria found fabricated explicit images of themselves circulating online, made from photos lifted off social media. Victoria Police arrested a teenage boy in connection with the case. Australia's eSafety Commissioner has since treated synthetic image based abuse as a priority enforcement area.
Telegram deepfake rings targeting schools
In August 2024, South Korean journalists and activists exposed networks of Telegram chat rooms producing deepfake pornography of women and girls, including material organized by specific schools and universities. Lists of targeted institutions circulated publicly, and the government announced emergency measures. The episode showed how face swap porn scales when generation takes seconds and distribution is anonymous.
Face swap porn: the other half of the market
Face swap porn maps a real person's face onto someone else's body in existing sexual photos or videos. It differs from an undress app in technique, not in outcome: a real identity is attached to sexual content that person never made.
The January 2023 scandal around streamer QTCinderella showed the mechanics in public. After a popular streamer was caught viewing a deepfake site that hosted fake explicit content of her and other female creators, her tearful response video was watched millions of times. The site in question later shut down, but copies appeared within weeks.
The largest hub of this ecosystem, MrDeepFakes, shut down in May 2025 after losing critical service providers, following years of pressure from researchers and platforms. Shutdowns help, but the Graphika data shows why the market persists: when one domain disappears, referral traffic moves to the next one within days.
What searches for "undress app" and "ai undresser" actually lead to
Search data shows a cluster of queries that funnel users toward this category of service:
According to lawsuits and security researchers, the services behind these queries tend to share the same traits:
- person_offAnonymous operators. The San Francisco complaint describes sites with no identifiable owner, no real company address and no channel for victims to request removal.
- paymentsPayment friction by design. Card and crypto payments with no refunds, aggressive upsells, and business models that rely on users being too embarrassed to dispute charges.
- cloud_uploadUnknown photo retention. Users upload photos of other people to servers with no verifiable privacy policy. What happens to those uploads afterward is unknowable by design.
- ads_clickAdvertising through mainstream platforms. Reporting by 404 Media in 2024 found undress app ads running on Meta platforms, which removed them only after journalists flagged them.
Risks for people who use an undress app
The risks of using these tools fall on the user as much as on the victim: fraud exposure, data theft, and prosecution in a growing list of jurisdictions.
Anonymous operators plus card payments plus zero refund policy is the classic profile of a scam storefront. Victims of the images are not the only people these sites exploit.
Uploading someone's photo means handing a stranger's biometric face data to an anonymous server. Your own payment details sit in the same database.
The Melbourne case ended with an arrest. The Spanish case went to juvenile court. The FBI warned in a June 2023 public service announcement that manipulated images from social media are used in sextortion schemes, and creating or sharing them can carry federal consequences.
If the person in the photo is under 18, the output is treated as child sexual abuse material in the US, UK, EU and Australia, synthetic or not. This is the single most prosecuted category of online crime, and "the AI made it" is not a defense investigators accept.
What to do if someone made a fake nude of you
Act in this order: preserve evidence first, then limit spread, then report. Removal tools exist and are free.
- Preserve evidence. Screenshot the content, copy URLs and usernames, note dates. Do this before reporting, because takedown deletes the evidence with the content.
- Stop the spread with hashing tools. If you are 18 or older, StopNCII.org creates a digital fingerprint of the image on your own device and shares only the hash with partner platforms, including Meta and TikTok, to block reuploads. If the original photo was taken before you turned 18, use Take It Down from the National Center for Missing & Exploited Children, which works the same way.
- Report to the platform hosting the content. Every major platform now has a dedicated category for non consensual intimate imagery, synthetic included.
- Request removal from Google Search through its non consensual explicit imagery removal form, so the content stops surfacing even if a copy stays online.
- Report to police. Bring the preserved evidence. In the US, the Cyber Civil Rights Initiative runs a crisis helpline at 1-844-878-2274. Elsewhere, ask for the unit handling image based abuse.
FAQ
What is an AI undress app?expand_more
An AI undress app is a tool that takes a normal clothed photo of a real person and uses generative AI to fabricate a fake nude or sexualized image of that person. The output is synthetic, but it is created without consent and is treated in many jurisdictions as image based abuse.
Are AI undress apps legal?expand_more
In many countries and US states, creating or sharing fake sexual images of real people without consent is a crime, and users have already faced arrests, school discipline and civil lawsuits. When the person in the photo is a minor, the output is treated as child sexual abuse material, one of the most serious categories of online crime.
Do undress apps show real photos?expand_more
No. An undress app cannot see through clothing. It invents a synthetic body that never existed. The harm, however, is real: viewers often believe the image, and victims report the same shame, harassment and reputational damage as if a real photo had leaked.
Can an undress app steal my photos or card data?expand_more
Yes, that risk is documented. Most of these services are run anonymously, publish no verifiable privacy policy, and require card or crypto payments. Investigators and city lawsuits describe operators with no accountability, unclear photo retention, and business models built on charge disputes and data harvesting.
What is face swap porn?expand_more
Face swap porn is a type of deepfake where a real person's face is mapped onto someone else's body in existing sexual photos or videos. Unlike undress apps, which fabricate a body from a clothed photo, face swapping hijacks existing explicit content. Research from Security Hero found that about 98 percent of deepfake videos online are pornographic, and about 99 percent of the people targeted are women.
What should I do if someone made a fake nude of me?expand_more
Save evidence (links, screenshots, usernames) before anything is deleted, report the content to the platform, and create a digital fingerprint of the image with StopNCII.org if you are 18 or older, or with Take It Down from NCMEC if you were under 18 when the original photo was taken. Then file a removal request with Google and report the case to local police.
Can undress apps be used on photos of minors?expand_more
Technically yes, and it has already happened in documented school cases in Spain, the United States and Australia. Legally, the output is child sexual abuse material regardless of being synthetic. The Internet Watch Foundation has confirmed a growing number of reports involving AI generated child sexual abuse imagery.
Why do undress apps keep appearing if they cause harm?expand_more
Because they are cheap to run and profitable. Graphika counted about 24 million visits to 34 undressing services in one month of 2023, and a San Francisco lawsuit reported more than 200 million visits to 16 such sites in the first half of 2024. Anonymous operators relaunch under new domains as soon as payment processors or hosts cut them off.
Sources and further reading
- linkGraphika, "A Revealing Picture" report on synthetic NCII providers, December 2023 (graphika.com).
- linkSecurity Hero, "The State of Deepfakes: Landscape, Threats, and Impact," 2023 (securityhero.io).
- linkSensity, deepfake census reporting, 2019 (sensity.ai).
- linkSan Francisco City Attorney David Chiu, lawsuit against 16 "undress" websites, August 2024 (sfcityattorney.org).
- linkFBI Internet Crime Complaint Center, public service announcement on sextortion with manipulated images, June 2023 (ic3.gov).
- link404 Media, reporting on undress app advertising on Meta platforms, 2024 (404media.co).
- linkInternet Watch Foundation, reporting on AI generated child sexual abuse imagery, 2024 (iwf.org.uk).
- linkStopNCII.org (adults) and Take It Down by NCMEC (minors), image hashing tools (stopncii.org, takeitdown.ncmec.org).
- linkPress coverage of the Almendralejo case (September 2023), Westfield High School case (October 2023), Bacchus Marsh Grammar case (June 2024), and South Korea Telegram investigations (August 2024).