This is a research prototype. The data and analyses are preliminary and not yet validated — we'd welcome your .
Federal prosecutors in New York have charged two individuals with violating the TAKE IT DOWN Act by creating and distributing AI-generated deepfake pornography. The defendants allegedly used generative AI to transform non-explicit source images of women—including celebrities, politicians, and private citizens—into sexually explicit content. This case represents one of the first major federal prosecutions under the new law, which criminalizes the nonconsensual publication of AI-generated digital forgeries.
DOJ alleged that Texas defendant Arturo Hernandez published about 113 albums of AI-generated deepfake pornography depicting about 50 identifiable women, including non-public figures and recent high school graduates, during a period anchored to May 2025. Prosecutors said non-explicit source images were allegedly transformed into sexualized depictions. The case was among the first prosecutions announced under the TAKE IT DOWN Act.
Domain classification, causal taxonomy, severity scores, and national security assessments were LLM-classified and may contain errors.
Using AI systems to gain a personal advantage over others such as through cheating, fraud, scams, blackmail or targeted manipulation of beliefs or behavior. Examples include AI-facilitated plagiarism for research or education, impersonating a trusted or fake individual for illegitimate financial benefit, or creating humiliating or sexual imagery.
Also classified as
The generated nonconsensual deepfake pornography was published on public websites and viewed millions of times, exposing internet users to toxic and abusive content.
The defendants used generative AI to create humiliating and sexually explicit imagery of specific, identifiable victims without their consent for distribution on adult platforms.
Human
Due to a decision or action made by humans
Intentional
Due to an expected outcome from pursuing a goal
Post-deployment
Occurring after the AI model has been trained and deployed
The harm was caused by the intentional decisions and actions of human defendants who used deployed generative AI tools to create and distribute nonconsensual deepfake pornography.
Added categories peak at 2: Minor (Psychological Harm, Direct)
Federal prosecution of two U.S. citizens for generating and distributing nonconsensual deepfake pornography of 140 women, including political figures and celebrities, under the TAKE IT DOWN Act. While the incident caused severe psychological harm and privacy violations, its national security impact is minor, representing a domestic criminal matter managed by law enforcement.
The AI systems were used as tools by the defendants to perform specific, well-defined functions (morphing non-explicit source images into explicit deepfakes) based on direct user inputs.
Limited Risk: The incident involves AI-generated content such as deepfakes, which are subject to specific transparency obligations under the EU AI Act to ensure users are aware they are interacting with synthetic media.
The primary function of the AI systems used in this incident was to generate realistic deepfake images and videos of identifiable individuals in sexually explicit scenarios.
The reports provide clear, detailed factual accounts of the arrests, the specific charges, the number of albums and victims involved, and the legal context surrounding the TAKE IT DOWN Act.
None
Technology
Sophisticated Generative AI Tools
Sophisticated tools allow easy generation of explicit deepfakes.
Inadequate Image Morphing Guards
AI tools allowed morphing clean photos into explicit images.
Data Inputs
Unauthorized Use of Real Photos
Defendants used real photos of victims to train or prompt AI.
Access to Public Images
Publicly available images of victims were exploited by AI tools.
Human Factors
Predatory Intent of Creators
Defendants deliberately created deepfakes to degrade victims.
Lack of Respect for Consent
Creators ignored victim consent when publishing intimate media.
Process and Methods
Inadequate Platform Moderation
Platforms failed to filter and block deepfake content uploads.
Easy Online Distribution
Websites allowed easy sharing and viewing of explicit material.
Management
Lack of Platform Safeguards
Platform management failed to establish robust content removal policies.
Regulatory Environment
Historical Regulatory Gaps
Lack of strict laws historically allowed deepfake spread before new acts.
Platform Compliance Failures
Some platforms failed to implement active removal processes.
AI system involved
An AI system was involved, but the reports do not identify it.
Identified by automated extraction; can differ from the alleged developers above.
