This is a research prototype. The data and analyses are preliminary and not yet validated — we'd welcome your .
A woman in St. Petersburg, Florida, falsely reported a burglary and sexual battery, using an AI-generated image of a nonexistent man created with ChatGPT. The image, part of a viral TikTok trend, was found in a deleted folder predating the alleged incident. The false report led to a significant police response, diverting resources. The woman later pleaded no contest to misdemeanor false reporting.
A St. Petersburg, Florida woman, Brooke Schinault, used a ChatGPT-generated image of a nonexistent man while falsely reporting that he broke into her home and sexually battered her. Police said the image resembled a viral "AI homeless man" prank and was found in a deleted folder dated before the alleged attack. The report prompted officers, rescue personnel, a detective, and forensic DNA collection. She later pleaded no contest and was adjudged guilty of misdemeanor false reporting.
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 AI was used to generate a false image that was presented as real evidence to mislead law enforcement.
The user employed ChatGPT to generate a synthetic image to deceive law enforcement and fabricate a violent crime for personal attention.
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 incident was caused by a human user intentionally generating a fake image using ChatGPT to perpetrate a hoax and falsely report a crime.
An individual used ChatGPT to generate a fake image of a suspect to support a false police report of a violent crime. While the incident diverted local emergency response resources, it was a localized criminal hoax with negligible national security implications.
The report describes the user typing a prompt into ChatGPT to generate a specific image of a homeless man, which she then used to fake a crime. The AI functioned as a simple tool executing a direct generation command.
Risk Level 3: Limited Risk. The system involved is ChatGPT, a general-purpose chatbot used to generate synthetic media (an AI-generated image), which falls under transparency obligations for AI-generated content.
The user utilized ChatGPT's image generation capabilities to create a synthetic image of a homeless man to insert into a photo of her home.
The reports provide clear, consistent, and detailed accounts of the incident, including the specific AI tool used (ChatGPT), the user's actions, the police response, and the legal outcome. There is no conflicting information regarding the core facts of the AI's role.
None
Technology
Realistic Image Synthesis
AI tools allowed generating convincing synthetic humans in real environments.
Absence of Provenance Metadata
Generated image lacked digital watermarks to easily identify it as synthetic.
Data Inputs
User Prompting of Deceptive Scenes
AI accepted prompts to insert a stranger onto a photo of a private couch.
Use of Real Residential Photos
System allowed uploading personal home photos to overlay AI-generated figures.
Human Factors
Desire for Attention and Hoaxing
User manipulated AI to fabricate a crime due to mental health struggles.
Viral Social Media Influence
TikTok trend normalized generating fake intruder photos to prank others.
Process and Methods
Weak Content Guardrails
AI platforms lacked filters to prevent generation of deceptive crime evidence.
Management
Inadequate Abuse Risk Assessment
Developers failed to mitigate use of AI images for false police reports.
Regulatory Environment
Lack of Synthetic Media Regulation
No legal framework mandated labeling or restricted realistic fake generation.
AI system involved
Identified by automated extraction; can differ from the alleged developers above.
