Schools are Facing a New Cyberbullying Challenge: AI-Generated Explicit Content

As generative AI gets cheaper and easier to use, impersonation and disinformation aren’t just hitting banks, CEOs, or election cycles. A quieter wave is spreading through schools: students using AI to create non-consensual explicit images of classmates and teachers.

This isn’t “kids being kids.” It’s harassment, amplified by tools that can generate a convincing fake in minutes, and share it in seconds.

1. The Scale is Rising Rapidly

Reports involving AI-generated child exploitation imagery jumped from 4,700 in 2023 to over 440,000 in the first half of 2025.

A Thorn survey found 1 in 8 U.S. teens personally knows a minor who has been targeted by deepfake explicit images.

Deepfake abuse is also highly gendered, 99% of non-consensual explicit deepfakes depict women.

2. This is Happening Across Countries

• United States: In Florida, two middle school boys were charged after allegedly creating AI-generated explicit images of classmates.
• United States: An elementary teacher in Austin faced allegations involving AI-generated explicit images of students, alongside child abuse charges.
• United Kingdom: A recent Guardian investigation describes deepfake incidents in schools, including a case where one girl was so distressed she vomited.
• Turkiye: Turkish media reported a case in İzmir where a student allegedly used AI to create explicitimages of female students and teachers.

3. Why Deepfake Bullying is Different

As with many forms of harassment, victims often don’t report what happened because they fear blame, punishment, or not being believed. Deepfakes make that silence easier to exploit: perpetrators can insist “it’s not real,” while victims still face real-world consequences, humiliation, isolation, anxiety, and damage to reputation.

And because this form of abuse is still new for many communities, many schools lack clear policies, evidence-handling processes, and staff training to respond quickly and safely.

4. Being Careful Online is not Enough

Telling kids to “watch what you post” misses the point: deepfakes can be created from ordinary photos or even yearbook images. In many cases, what matters most is whether a student feels safe enough to speak up early.

That sense of safety often starts with how adults respond. When a child believes they’ll be blamed or punished, silence becomes the easiest option. But when they expect support they’re more likely to ask for help before the situation spreads.

5. Measures that Schools Can Take

Schools are catching up to a form of harm that moves faster than traditional response systems. This isn’t always treated like standard bullying, because synthetic explicit content can spread quickly and leave lasting damage. The most effective approaches tend to combine clear policy, trusted reporting, and staff readiness.

• Clear Classification and Pathways: Many schools find it helpful to define non-consensual explicit deepfakes as harassment and route cases through protection processes, not only discipline.
• A Trusted Reporting Route for Students: A simple, well-communicated reporting path can make a real difference. That often includes:

– one or two clearly named staff contacts

– the ability to report confidentially (and, where possible, anonymously)

– reassurance that students will be supported and taken seriously

• Updated Codes of Conduct for the AI Era: Policies are increasingly being updated to explicitly prohibit:

– generating synthetic explicit content involving anyone in the school community

– sharing, storing, or threatening to share it

– pressuring others to create or distribute it

Clear consequences matter, but so does clarity, students need to know exactly what crosses the line.

• Student Education That Doesn’t Normalize Harm: Short, age-appropriate sessions can focus on:

– why “it’s fake” doesn’t reduce impact

– how to report safely

– bystander responsibility (not forwarding, not joking, not amplifying)

• Staff Training: Teachers and administrators benefit from practical guidance on:

– evidence handling and documentation

– mandatory reporting thresholds and escalation routes

– trauma-informed language that avoids retraumatizing prompts (for example, asking students to “prove it”)

6. Measures that Government Can Take

• Update Laws for Synthetic Abuse: In many jurisdictions, “synthetic” content still falls into legal gray areas. Clearer frameworks can help ensure that legislation covers:

– the creation and distribution of non-consensual explicit synthetic media

– stronger protections for minors, including faster intervention mechanisms

– accountability for adults in positions of trust (schools, clubs, coaching)

• Enable Faster Removal: Rapid removal of non-consensual explicit deepfakes can help limit further spread and reduce harm, especially in school settings where content can circulate quickly.
• Fund School Capacity, not Only Awareness: Awareness campaigns help, but schools also need operational capacity. Funding priorities include staff training and guidance, reporting and student support services, access to digital forensics support for serious cases and counseling capacity for affected students.
• Expand Victim Support and Legal Help Pathways: Families often need a clear route to support, including practical guidance on where and how to report, help navigating takedown processes and access to counseling and legal support.
• Push for “Safety by Design”: Require risk controls in high-risk tools and services (especially those facilitating sexualized misuse), alongside privacy protections and transparency about enforcement.

7. Evidence-Based Verification

Cyberette helps teams verify media and detect manipulation so institutions can respond with evidence, not uncertainty. Our platform can indicate where and when manipulation occurred, supported by clear forensic reporting.

Have questions? Reach out at info@cyberette.ai