This is a research prototype. The data and analyses are preliminary and not yet validated — we'd welcome your .
Dozens of Dutch female public figures, including politicians, athletes, and artists, were targeted with non-consensual deepfake pornographic videos created using AI and hosted on an offshore platform.
Dutch reporting in March 2024 identified roughly 70–80 prominent women, including politicians, presenters, athletes, celebrities, and members of the royal family, as reportedly depicted in pornographic deepfake videos made by using AI to place their faces on performers' bodies. Some videos had more than 100,000 views. Several women said they had not known the videos existed and filed or planned police reports.
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 synthetic pornographic videos expose internet users and the victims themselves to highly toxic, non-consensual sexual content.
The incident involves the creation of humiliating sexual imagery targeting specific female public figures for targeted humiliation and financial gain.
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 human actors intentionally using deployed AI deepfake generation tools to create and distribute non-consensual pornographic videos of specific victims.
Added categories peak at 2: Minor (Psychological Harm, Direct)
Dozens of Dutch female public figures, including MPs and ministers, were targeted with AI-generated non-consensual deepfake pornography hosted on an offshore platform. While not a state-sponsored national security threat, the incident highlights how AI-enabled harassment can target state leadership, threaten democratic participation, and challenge law enforcement capabilities.
The reports describe users using AI-powered apps and algorithms to manipulate videos by pasting faces onto other bodies. This fits the 'Tool' classification as the AI performs a specific, well-defined function based on direct human input.
Limited Risk: The report describes the creation of deepfakes, which are explicitly classified under the EU AI Act as posing a moderate risk requiring transparency obligations.
The AI systems were used specifically to generate manipulated pornographic videos (deepfakes) by pasting the faces of victims onto other bodies.
The reports are highly detailed, consistent, and come from reputable Dutch news outlets. They explicitly identify the victims, the platform, the technology used, and the societal/legal responses.
None
Technology
Advanced AI Face Swapping
AI algorithms convincingly paste faces onto explicit video bodies.
Proliferation of Deepfake Apps
Easy-to-use applications allow anyone to generate realistic deepfakes.
Data Inputs
Publicly Accessible Photos
Abundant online images of public figures provide training data for AI.
Human Factors
Uploader Financial Motivation
Creators are incentivized by platform payouts tied to video view counts.
Audience Demand and Sharing
Viewer consumption and requests drive the creation of new deepfakes.
Dismissal of Real Harm
Perpetrators justify deepfakes as harmless fantasy rather than abuse.
Process and Methods
Slow Law Enforcement Response
Victims face months of police silence after filing official complaints.
Ineffective Takedown Procedures
Lack of rapid administrative orders prevents quick removal of videos.
Advice to Remain Silent
Initial advice to stay quiet to avoid driving traffic delayed action.
Management
Platform Anonymity Policies
Platforms advise uploaders to stay anonymous to avoid legal consequences.
Monetization of Exploitation
Payment processors and sites historically facilitated deepfake monetization.
Regulatory Environment
Offshore Hosting Havens
Platforms register in jurisdictions like Nevis to evade legal prosecution.
Inadequate Law Enforcement
Existing laws against deepfakes are not actively or quickly enforced.
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.
