Artificial intelligence deepfakes have transformed school bullying from cruel messages into a crisis involving fabricated videos, fake audio, and sexually explicit images created without consent. A nationally representative survey of K-12 principals found that 13 percent reported bullying incidents involving AI-generated deepfakes during the 2023-24 or 2024-25 school years, with secondary schools hit hardest. Twenty percent of middle school principals and 22 percent of high school principals said their schools had experienced bullying involving AI-generated images, fabricated photographs, deepfakes, or false online identities.

The problem has moved beyond entertainment and political misinformation into classrooms where easily accessible artificial intelligence tools can be weaponized in seconds. A student with nothing more than a yearbook photo and access to a free or inexpensive app can create convincing fabrications that spread through group chats, anonymous accounts, and gaming communities before administrators even know a crisis is unfolding.

What Makes AI Deepfakes Different From Traditional Cyberbullying?

Deepfake technology adds a dangerous dimension to school harassment that traditional bullying never had: the ability to manufacture convincing visual or audio evidence of events that never occurred. A deepfake is an image, video, or audio recording created or manipulated with artificial intelligence to imitate a real person. It can place someone's face into a fabricated scene, recreate a person's voice, or make someone appear to say or do something that never happened.

The technical barrier to creating these fakes has collapsed. Synthetic media has existed for years, but generative AI has dramatically lowered the cost and skill required. Creating a convincing fabrication once required editing experience, specialized software, and significant time. Today, a student may need little more than a publicly available photograph and access to an inexpensive or free AI-powered app. A student acting impulsively may create an image in seconds without fully considering the emotional, disciplinary, or legal consequences.

What begins as an attempt to embarrass one classmate can quickly become a schoolwide crisis when the content is shared through group messages, anonymous accounts, social media platforms, or gaming communities. Classes are disrupted. Investigations begin. Students gather in hallways to discuss what they have seen. Screenshots move from one platform to another. Employees face questions about conduct that never occurred, while targeted students are forced to defend themselves against images they never created.

How Are Students Being Targeted Without Sharing Anything Explicit?

Some of the most disturbing school-related cases involve "nudify" tools, which use artificial intelligence to transform an ordinary photograph into a fabricated nude or sexually explicit image. A student does not have to send an intimate photograph to become a victim. A yearbook portrait, athletic photograph, social media post, or picture taken in a school hallway can be enough to create exploitative content. The image may then be circulated among classmates, posted to an anonymous account, or used to threaten and control the person depicted.

The National Center for Missing and Exploited Children has warned that generative AI is being used to create exploitative imagery involving children, identifying peer victimization as a growing child safety concern. Girls have been disproportionately targeted in many reported incidents, although any student can become a victim. The targeted student may not learn that an image exists until a friend sends a screenshot or rumors begin circulating at school.

The discovery can create an immediate sense of panic and helplessness. Even when classmates know an image has been fabricated, they may continue sharing it for attention, entertainment, or cruelty. Other students may believe it is authentic. The victim is then placed in the painful position of repeatedly explaining that the image is not real while knowing that peers have viewed, saved, discussed, and possibly redistributed it.

The resulting harm can extend throughout a student's life at school. A targeted student may avoid classrooms, lunch periods, athletics, extracurricular activities, or social events. Attendance and academic performance may decline. Some students experience anxiety, depression, humiliation, isolation, or fear that the image will resurface during college admissions, employment searches, or future relationships. Describing the content as artificial should never be used to minimize the experience of the person depicted. The technology may have fabricated the image, but it cannot erase the invasion of privacy, loss of control, reputational damage, and psychological distress that follow.

How Schools Can Respond to Deepfake Incidents

• Establish Clear Reporting Channels: Create multiple ways for students to report deepfake content, including anonymous reporting options, so victims feel safe coming forward without fear of blame or punishment.
• Develop Verification Protocols: Train staff to verify whether content is authentic before launching investigations or taking disciplinary action, preventing false accusations based on fabricated evidence.
• Provide Trauma-Informed Support: Offer counseling and mental health resources to targeted students, recognizing that the psychological impact of deepfake victimization can be severe and long-lasting.
• Implement AI Literacy Training: Educate students about how deepfakes are created, the legal consequences of creating them, and the real harm they cause to victims.
• Create Sextortion Prevention Education: Teach students about the warning signs of sextortion attempts and encourage them to report threats immediately to trusted adults rather than complying with demands.

The Sextortion Threat: When Deepfakes Become Extortion

Deepfake technology also increases the threat of sextortion, in which an offender uses sexual content or the threat of exposure to demand money, additional images, or other forms of compliance. Traditionally, sextortion schemes have often begun when a young person sends an intimate photograph to someone they believe they can trust. With generative AI, an offender may not need an authentic image. A fabricated but convincing picture can create enough fear and shame to manipulate the victim.

An offender could take a photograph from a student's social media account, alter it, and threaten to send the result to parents, classmates, teammates, teachers, or college recruiters. The student may comply because the potential consequences feel immediate, regardless of whether the content is authentic. Young people can be particularly vulnerable to this form of coercion. A student who fears punishment or public embarrassment may send money, provide additional content, or attempt to handle the situation alone rather than approaching an adult.

That isolation gives the offender more control. School communities must make it clear that students who are threatened or exploited will receive help rather than blame. They should be encouraged to stop communicating with the offender, avoid sending money or additional images, preserve available evidence, and contact a trusted adult immediately. The FBI identifies sextortion as a serious crime and urges students, families, educators, and caregivers to understand the warning signs.

Teachers and Administrators Are Targets Too

Students are not the only people at risk. Teachers, principals, coaches, superintendents, and school board members can also become deepfake targets. A student can manipulate a teacher's photograph to create an inappropriate image. AI-generated audio can make it sound as though a coach threatened a player or a principal made a discriminatory statement. A fabricated video can falsely depict an employee engaging in misconduct.

Such content can be especially damaging in a school environment where trust and reputation are essential to effective leadership and instruction. An administrator arriving to learn that a fabricated video involving a teacher has circulated through student group chats overnight faces an immediate crisis. Families begin calling the district about an alarming audio recording attributed to the superintendent, while administrators scramble to determine whether the voice is authentic. None of the content is real, yet every person affected must deal with real consequences.

This is not simply another violation of a school device policy. It is a rapidly developing school safety issue involving cybersecurity, cyberbullying, student mental health, sexual harassment, privacy, public trust, and crisis communication. Schools must recognize that deepfakes represent a fundamentally new threat that requires new responses, new training, and new partnerships with law enforcement and technology experts.