The next great safeguarding challenge is already in our classrooms

AI is changing the digital world children inhabit. School safeguarding must evolve with it.

Schools have spent years developing their response to online safety. We teach pupils about cyberbullying, harmful content, grooming, privacy and the importance of questioning what they encounter online. Staff receive safeguarding training, policies are updated and concerns are escalated through established procedures. Artificial intelligence (AI) adds new dimensions to this work.

A child may now interact with a chatbot that appears understanding and emotionally responsive. They may encounter convincing images, audio or video that never existed. They may disclose personal information to a tool without understanding where that information goes or how it might be used. They may also turn to AI for advice about relationships, anxiety or personal difficulties. These experiences are not confined to a future digital world. They are part of the environment in which children are growing up. AI safeguarding is not a separate concern for the IT department. It is part of safeguarding children.

The risks are changing

Some AI-related risks are extensions of familiar online safety concerns. Others require schools to think differently. Deepfakes can make fabricated images or recordings appear authentic. AI-generated content can be persuasive even when it is inaccurate. Conversational systems can create an impression of understanding or authority that a young person may struggle to evaluate. The risks are not limited to deliberately harmful use.

A pupil may rely on an AI-generated response to a sensitive personal question because it is immediate, private and apparently non-judgemental. They may mistake a confident answer for a reliable one. That is why AI literacy and safeguarding are increasingly inseparable. Young people need to understand not only how to use AI, but how to recognise when its use may be unsafe, misleading or inappropriate.

Staff need more than a list of prohibited tools

A safeguarding response built entirely around banning particular applications will struggle to keep pace with technological change. Schools need clear rules about approved tools, data protection and the use of pupil information. But staff also need to understand the underlying risks. What should a teacher do if a pupil reports an AI-generated intimate image? How should a school respond when a child appears emotionally dependent on a chatbot? What information should never be entered into an unapproved AI system?

These questions need to be addressed through safeguarding leadership, staff development and clear reporting arrangements. They should not depend on individual staff members who happen to be knowledgeable about AI.

Education is part of protection

Children also need opportunities to discuss AI openly. If pupils believe that any admission of AI use will result in punishment, they may be less willing to disclose an uncomfortable or harmful experience. Schools should create a culture in which pupils can ask questions, report concerns and understand why particular boundaries exist.

This does not mean presenting AI as inherently dangerous. It means helping children recognise both its possibilities and its limitations. The same balanced approach should inform conversations with parents, who may be navigating unfamiliar technology themselves.

A question for every safeguarding leader

At AiEd, we believe every school should review whether its safeguarding arrangements reflect the AI-rich world children now inhabit. That means considering AI within policies, training, curriculum and reporting procedures, and ensuring that safeguarding leaders are supported to keep their knowledge current.

We should not wait for a serious incident before deciding that AI deserves attention. The question for schools is straightforward: do our safeguarding arrangements reflect the digital world children are actually experiencing? If the answer is not yet, this is the moment to act.

‍