This is a research prototype. The data and analyses are preliminary and not yet validated — we'd welcome your .
Radnor High School students were targeted by the creation and dissemination of AI-generated sexualized images, prompting a police investigation and harassment charges against a juvenile. The incident caused significant psychological distress among the student body and sparked community calls for updated school policies regarding AI misuse and student safety. Experts noted that while major AI platforms have safeguards, unregulated 'rogue' tools remain easily accessible for such malicious use.
Radnor High School in Radnor Township, Pennsylvania, and local police reportedly investigated a purportedly AI-generated video that allegedly depicted several students "in an inappropriate manner." The principal said affected families were contacted and offered support, and police were notified.
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 students and school community were exposed to toxic, sexually explicit synthetic media depicting minors.
The incident involved a juvenile using AI tools to create and disseminate humiliating, sexualized deepfake images of specific classmates, which constitutes targeted harassment and manipulation using synthetic media.
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 (a juvenile) intentionally using a deployed AI tool to generate and disseminate non-consenting sexualized images of classmates.
Added categories peak at 3: Substantial (Child Sexual Exploitation & Abuse (CSEA), Direct)
A localized incident involving a juvenile using unregulated AI tools to create and disseminate non-consensual sexualized deepfakes of classmates. While causing significant psychological distress and prompting local police and school policy responses, the national security impact remains minor, representing an established consumer-level threat rather than a systemic risk.
The AI was used as a tool by a human user to perform a specific, well-defined function (generating sexualized images from input photos) without autonomous planning or context retention. The report mentions 'rogue tools that allow criminals to engage them for a fee to do things like create sexualized images'.
Limited Risk: The incident involves AI-generated content such as deepfakes, which falls under Risk Level 3 due to transparency and disclosure concerns, though the malicious use itself is illegal.
The AI system was used to generate fake sexualized images and videos of students from innocent photos.
The reports provide clear, consistent details about the nature of the incident, the parties involved, the legal actions taken, and the community impact. While the specific AI tool used is not named, its functional category (unregulated deepfake generation tool) is clearly described by cybersecurity experts.
None. The incident is unambiguously a case of non-consenting deepfake harassment targeting minors.
Technology
Rogue Unregulated AI Apps
Unregulated tools allow creation of sexualized images for a fee.
Accessible Deepfake Tech
Powerful deepfake capabilities are now easily accessible in kids' hands.
Data Inputs
Public Social Media Photos
Publicly shared photos and selfies are weaponized to generate fake images.
Human Factors
Misuse of AI by Juveniles
A juvenile student deliberately used AI to generate inappropriate media.
Misplaced Trust in Peers
Students share photos with peers who turn out to be untrustworthy.
Process and Methods
Delayed Investigation Timelines
Parents criticized slow investigation speeds and lack of firm timelines.
Inadequate Communication
School used silence and minimization instead of transparent updates.
Management
Lack of Support for Victims
District lacked policies to properly support student victims of AI abuse.
No Parental Consent for Interviews
School interviewed students without parental consent during investigation.
Regulatory Environment
Lagging District Policies
Policies failed to keep pace with rapid technological advancements.
Unclear Off-Campus Jurisdiction
Uncertainty regarding when district rules apply to off-campus behavior.
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.
