For decades, the school photo served one purpose. It celebrated students, promoted the school and gave parents something to keep. That assumption no longer holds. The eSafety Commissioner’s latest Online Safety Advisory confirms that the weaponisation of school imagery through generative AI now ranks among the fastest-growing threats facing Australian schools. Photos posted on official school websites, newsletters and social media accounts are being harvested by bad actors. Some are strangers. Many are current or former students. The numbers behind this shift explain why school leaders across the country are acting fast.
More Than 100 Reports in Three Months
Between January and March, the eSafety Commissioner logged more than 100 reports of anonymous accounts targeting school staff and students. Most operated on TikTok and Instagram. Nearly all of them relied on imagery harvested from official school social media accounts or websites. That detail matters. The source material for this abuse rarely comes from private accounts or leaked images. It comes from content schools published themselves, in good faith, for their communities. The abuse takes two main forms.
The first targets staff. Attackers take official headshots of principals and teachers and produce humiliating face-swaps, AI-generated dance videos, fabricated ratings and false stories designed to damage reputations. The second is far more serious. Ordinary photos of students under 18 are being run through “nudify” apps that digitally remove clothing or generate explicit child sexual exploitation material. Reports of digitally altered intimate deepfakes involving minors have more than doubled over the past 18 months.
The Scale of the Nudify Industry
The tools driving this abuse are easy to find and often free. Research shows some carry no age restrictions at all. Two of the world’s most-visited AI nude image websites attracted about 100,000 visitors per month in Australia alone. eSafety identified both as services used to generate explicit deepfake images of students in Australian schools. Investigators found these services marketed features such as undressing “any girl,” with a “schoolgirl” option among the settings.
eSafety Commissioner Julie Inman Grant has described the technology as “putting potential online harm on steroids.” Her data shows at least one incident of deepfake image-based abuse takes place every week in Australian schools. The human cost is already visible. NSW Women’s Safety Commissioner Hannah Tonkin has spoken directly with high school students living with this fear. “Many are seeing it happen to their friends and they’re really scared it’s going to happen to them,” Dr. Tonkin said. She described nudify apps as “disgusting technology” designed to target, dehumanise and inherently degrade women. Enforcement is, however, producing results. As a result of eSafety action over the past year, seven of the most widely used nudify services in Australia have either withdrawn access or moved to comply with the Online Safety Act.
How Schools Are Responding
Following warnings amplified by 7NEWS and Nine News, schools are overhauling their digital footprints. Three practical changes stand out.
1. Stripping Faces From Public Media
Many campuses are removing identifiable student images from public websites. Replacements include silhouettes, back-facing shots and even AI-generated fictional students for promotional material.
At least one Canberra Catholic school removed all photographs from its website almost overnight, leaving a single image of an empty playground. Schools are now weighing options that would have seemed absurd two years ago, including empty playgrounds, the backs of children’s heads and paid models.
2. Locking Down Newsletters
Open public feeds are giving way to password-protected portals and, in some cases, printed newsletters. The goal is straightforward: community updates should reach families without becoming raw material for anonymous accounts.
3. Rewriting Consent Frameworks
Parental permission forms written for a pre-AI era are consequently being updated. Schools are ensuring families understand the specific digital risks before signing off on any photography. This step is commonly overlooked, yet it gives parents the informed choice they deserve.
For school leaders, the immediate practical step is to audit every public channel. Counting how many identifiable student and staff images sit on open feeds helps clarify which ones genuinely need to remain public.
Why This Extends Beyond Schools
The problem, furthermore, reaches past individual campuses. In 2024, Human Rights Watch documented that personal photos of Australian children, including images posted by schools, had been scraped and used to train AI models. The organisation stated that children should not have to worry that photographs shared online by their schools will be scraped, manipulated and turned into harmful content beyond their control. The global picture confirms Australia is part of a wider trend. A joint investigation by WIRED and Indicator identified reported deepfake sexual abuse incidents across roughly 90 schools worldwide, affecting more than 600 students. Since 2023, students in at least 28 countries have been caught using generative AI to create sexualised images of classmates.
What Comes Next
The measures Australian schools are taking now are necessary and overdue. They are, however, incomplete on their own. The tools behind this abuse evolve faster than consent forms and website audits. Regulators have shown that enforcement works, as seven major services have already been pulled back from the Australian market. Sustained pressure on platforms, clear obligations under the Online Safety Act and ongoing education for students and parents will determine whether schools stay ahead of the threat.
For educators, school leaders and policymakers, the immediate task is clear. Every publicly posted image should be treated as a potential input for abuse. Reducing the exposed surface area and keeping consent frameworks current are, therefore, the two most practical steps available right now. The school photo will survive this moment. Where it lives, and who can access it, has already changed for good.
