Deepfakes and Children: What Parents and Schools Should Know

Deepfakes — synthetic images, audio, or video generated by AI to make a real person appear to say or do something they did not — have moved from a fringe curiosity to a mainstream child safeguarding issue. UK schools, the police, and helplines now routinely deal with incidents involving deepfake imagery of pupils, sometimes generated using easily-downloaded apps that promise to 'undress' a photo or place a child's face on adult content. Parents and schools need a shared understanding of the risks, the law, and the response. What counts as a deepfake harming a child? The most common categories include sexual deepfakes (a child's face placed on adult bodies or pornographic content), humiliation deepfakes (making a child appear drunk, naked, violent, or saying offensive things), bullying deepfakes (used in group chats to harass), and identity-theft deepfakes (impersonating a child to deceive their friends or family). Audio deepfakes — voice cloning — are increasingly used for scams and bullying. All of these can be created from photos a child has shared publicly, sometimes from a single image. The law in England and Wales is clear and getting stronger. Creating or sharing intimate or sexual deepfake imagery of a person without their consent is a criminal offence. Where the subject is under 18, the imagery is treated under existing child sexual abuse imagery laws — there is no 'AI loophole'. Threatening to share intimate imagery is also a separate offence. The Online Safety Act places duties on platforms to remove such material quickly, and the Internet Watch Foundation can issue takedown notices. School incidents follow a pattern. Often, a pupil uses a 'nudify' app on photos taken from social media, then shares the result in a group chat. The image spreads quickly, sometimes reaching the target child only after many peers have seen it. Schools should treat these incidents as serious sexual harm, not as 'banter'. KCSIE 2025 expects schools to consider whether to escalate to police, social care, and parents, to support the targeted child, and to address the behaviour of the perpetrator through a graduated response that takes their age and any safeguarding concerns into account. For parents, prevention has three parts. First, reduce the supply of usable photos. Children with public social media accounts and tagged images are easier to target; review privacy settings together. Second, talk to your child early about deepfakes — explain that no one's face is safe from this technology and that being a target says nothing about the victim. Third, agree how they will tell you if it happens. Many children fear being blamed for posting photos in the first place; reassure them that you will help, not punish. If an incident happens, the response is: support the child first, evidence second, report third. Reassure them, screenshot the imagery and the surrounding context without sharing it, then report to the school (if peers are involved), the platform, and — for sexual deepfakes of under-18s — the Internet Watch Foundation and the police. The Revenge Porn Helpline can support teenagers (and adults) where intimate imagery has been weaponised against them. Keep records of every report and ask for incident reference numbers. Finally, push for action. Ask your school how it handles deepfake incidents. Ask platforms why nudify-style services remain accessible. The more pressure on tools, platforms, and policymakers, the smaller the supply of harm.