Purported Deepfake Explicit Images of Middle School Students Allegedly Created and Circulated Using Mobile App in Goffstown, New Hampshire

In Goffstown, New Hampshire, unnamed eighth-grade boys at Mountain View Middle School allegedly used an AI-enabled app to create and circulate sexualized deepfake images and videos depicting female classmates.

Risk classification

• Primary risk domain: 4 Malicious actors
• Primary risk subdomain: 4.3 Fraud, scams, and targeted manipulation

The incident involves the targeted creation of humiliating and sexual synthetic imagery of specific victims for personal amusement or harassment, which aligns with the boundary rule B1 for targeted exploitation of public likeness.

Additional risk subdomains

• 1.2 Exposure to toxic content: The AI system was used to generate and distribute sexually explicit, abusive, and inappropriate content depicting minors among middle school students.

Causal factors

• Entity: Human
• Intent: Intentional
• Timing: Post-deployment

The harm was caused by the intentional decision of human users (the students) to use a deployed AI application to generate and distribute non-consensual explicit deepfakes of their classmates.

EU AI Act risk tier

• Risk tier: 3 Limited Risk

Limited Risk: The system is used to generate deepfakes, which are subject to specific transparency obligations under the EU AI Act to ensure users are aware the content is synthetic.

AI system and alleged parties

• AI system: None named
• AI purpose: Deepfake Image Generation; Deepfake Video Generation
• Behaviour type: Tool
• Alleged developer: Unknown image generator developers, Unknown deepfake technology developers
• Alleged deployer: Unnamed students at Mountain View Middle School
• Alleged harmed parties: Unnamed students at Mountain View Middle School, students, minors, Families of unnamed students at Mountain View Middle School, Epistemic integrity

Harm severity

Severity ratings are based on this scale.

Highest direct severity in any category: Substantial. Severity is scored from Negligible to Catastrophic in each harm category, for harm the reports describe as caused directly or indirectly by the AI system.

• Physical: direct Negligible, indirect Negligible
• Infrastructure: direct Negligible, indirect Negligible
• Property: direct Negligible, indirect Negligible
• Financial: direct Negligible, indirect Negligible
• Environmental: direct Negligible, indirect Negligible
• Malicious content: direct Minor, indirect Negligible
• Differential treatment: direct Negligible, indirect Negligible
• Civil rights: direct Negligible, indirect Negligible
• Democracy: direct Negligible, indirect Negligible
• Privacy: direct Minor, indirect Negligible
• Psychological: direct Minor, indirect Minor
• Epistemic: direct Negligible, indirect Negligible
• Child sexual exploitation and abuse: direct Substantial, indirect Negligible

Malicious content

Reported: Yes

Directly caused: The AI app was used to generate deepfake naked photos and videos of three minor girls, which were then shared among classmates.

Inferred additional harm: The explicit digital content may continue to exist on local devices or online platforms, posing an ongoing risk of further exposure and harassment.

Privacy

Reported: Yes

Directly caused: The victims' personal photos were processed without consent to generate highly sensitive, non-consensual explicit imagery, violating their bodily and personal privacy.

Psychological

Reported: Yes

Directly caused: The three female victims experienced severe emotional distress, anxiety, and fear of attending school, with one mother describing the impact as 'devastating' and another noting her daughter is 'very nervous to face these peers again.'

Indirectly caused: The parents of the victims experienced significant distress, anger, and frustration due to the incident and the school's delayed response.

Inferred additional harm: The victims are likely to suffer from long-term psychological trauma, social anxiety, and academic disruption as a result of the targeted harassment.

Child sexual exploitation and abuse

Reported: Yes

Directly caused: The incident involved the creation and distribution of AI-generated sexually explicit media depicting three middle school girls who are minors.

People affected

• Occurrences reported: 1
• People reportedly harmed: 3
• People reportedly exposed: 3

Potential causes

Management

• Prioritizing Legal Confidentiality: Prioritizing legal confidentiality delayed transparent communication by school.
• Inadequate School Incident Response: The school district lacked a rapid, proactive response protocol for AI abuse.

Technology

• AI Undressing App Accessibility: Easy-to-use AI apps allow users to easily generate non-consensual deepfakes.
• Low Barrier to Image Generation: AI requires only a single standard photograph to generate realistic deepfakes.

Data Inputs

• Unrestricted Personal Photos: Standard, easily accessible photos of classmates were used as inputs to the AI.

Human Factors

• Lack of Student Empathy: Students failed to comprehend the severe emotional impact of sharing deepfakes.
• Peer Pressure and Sharing Culture: Students shared the generated explicit content amongst their peer group.
• Lack of Parental Phone Monitoring: Parents did not monitor phone usage or apps installed by their children.

Process and Methods

• Delayed Parent Notification: The school district took a month to notify parents about the AI incident.
• Lack of School AI Education: Schools lacked established programs to educate students on AI dangers.

Regulatory Environment

• Lack of AI App Regulations: Absence of regulations governing the creation and distribution of AI apps.

Information quality

• Classification confidence: High
• Reason for confidence: The report provides clear, direct accounts from the victims' parents and details the specific nature of the AI abuse, the age of the students, and the school's response. The facts are consistent and sufficient to classify the incident with high confidence.
• Ambiguities identified: The specific name of the AI application used is not mentioned in the report.
• Alternative interpretations: None. The incident is clearly a case of targeted harassment and non-consensual deepfake generation involving minors.

Middle school students in New Hampshire used an AI-powered app to generate and distribute non-consensual explicit deepfakes of classmates. While causing severe localized psychological harm and violating the victims' privacy, the incident has negligible national security implications and is manageable through local educational and law enforcement procedures.

• Overall national security impact: Minor
• Scope: Single nation
• Primary target: United States
• Alleged perpetrator: Group

Threat characteristics

• Imminence: Long-term. The specific incident has already occurred and is being handled locally, though the proliferation of deepfake technology remains an ongoing strategic concern.
• Autonomy: Human-controlled. The AI application acted as a tool directly controlled and initiated by human users to perform a specific image-manipulation task.
• Novelty: Established threat. The use of AI-powered nudifying applications to generate non-consensual explicit imagery of individuals is an established and recurring threat.

Impact by dimension

• Physical security: Negligible. No physical systems, critical infrastructure, or kinetic targeting capabilities were involved or affected by this incident.
• Information security: Negligible. The incident involved localized peer-to-peer harassment and did not compromise intelligence capabilities or constitute a systematic information warfare campaign.
• Sovereignty: Negligible. No sovereign state authority, electoral systems, or core government decision-making processes were impacted or disrupted.
• Economic security: Negligible. There was no impact on financial systems, critical supply chains, or strategic technological competitive advantages.
• Societal stability: Minor. The incident represents a serious violation of privacy and rights for the minor victims, but the impact is localized to a single school and manageable through standard local procedures.