This is a research prototype. The data and analyses are preliminary and not yet validated — we'd welcome your .
A 23-year-old woman in Kyrgyzstan became the victim of a sexualized deepfake generated and distributed by her classmate using an AI Telegram bot, causing her severe psychological distress and reputational damage.
A 23-year-old woman in Kyrgyzstan said a former classmate used a Telegram AI bot to create a nude image from her Instagram photo without consent; it reached his contacts. She reported reputational and psychological harm. He acknowledged bot use but attributed circulation to a hack of his Telegram account. Creation timing and responsibility for sharing remain disputed.
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 nude deepfake was distributed to multiple contacts, exposing them to non-consensual sexual content.
The incident involves the creation of humiliating sexual imagery targeting a specific victim to damage her reputation, which fits under targeted manipulation and abuse.
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 a human decision to use an AI tool to generate a non-consensual sexualized image of a classmate, which was then distributed.
Added categories peak at 2: Minor (Psychological Harm, Direct)
An individual in Kyrgyzstan used a public AI Telegram bot to create and distribute a non-consensual sexualized deepfake of a classmate. While causing severe personal and psychological harm to the victim, the incident remains a localized criminal and harassment issue with negligible direct implications for national security.
The AI system is a specialized Telegram bot used for image processing and deepfake generation. It operates as a tool, performing a specific function on direct input (the victim's photo) to generate a deterministic output.
Limited Risk: The system is an AI-generated content tool used to create deepfakes, which carries specific transparency obligations under the EU AI Act.
The AI system was a Telegram bot specifically used to generate a sexualized deepfake by manipulating a real photo of the victim.
The report provides detailed first-hand accounts from both the victim and the accused, along with statements from the police and legal counsel. While there is a dispute over whether the distribution was due to a hack or intentional action, the core facts of the AI's use to generate the deepfake and the resulting harm are clear.
The distribution of the image could have been a malicious hack targeting the perpetrator rather than a direct intentional act of distribution by him, though he admits to generating the image.
Technology
AI Nudification Bots
Telegram bots easily generate realistic non-consensual deepfake porn.
Realistic Image Generation
Deepfake was highly realistic, making it difficult to prove it was fake.
Data Inputs
Public Social Media Photos
The source photo was easily downloaded from a public Instagram account.
Human Factors
Intentional Misuse of AI
User deliberately used AI to strip classmate's photo for amusement.
Phishing Vulnerability
User entered account confirmation code into a malicious Telegram bot.
Process and Methods
Flawed Technical Investigation
Investigators lacked AI literacy, confusing ChatGPT with Telegram bots.
Delayed Forensic Analysis
Police failed to conduct timely technical audit of devices and accounts.
Regulatory Environment
Weak Deepfake Regulations
Ambiguous laws fail to deter and properly penalize deepfake creation.
Lack of Platform Control
Telegram fails to proactively ban automated AI nudification bots.
