Most discussion of pupils using AI for homework focuses on cheating. The safeguarding questions matter more, and get far less attention. Here's what schools should actually be alert to.

Beyond the academic honesty question

Whether AI use is cheating is a real question — but it isn't a safeguarding one. These are the risks that belong on a DSL's radar.

The six risks worth knowing

1. Exposure to harmful content

Guardrails on consumer AI vary and can be circumvented. A pupil researching a sensitive topic for homework may encounter material that is inaccurate, distressing or inappropriate, without the filtering your school network would apply.

2. Personal data disclosure

Pupils routinely paste in personal information — their name, school, address, details about family — without any sense that it's going to a commercial third party. Younger pupils in particular treat a chatbot as private.

3. Disclosure of distress to a bot

The most serious. A child who mentions self-harm, abuse or crisis to an AI while doing homework has made a disclosure to something that cannot safeguard them and will not tell anyone. See AI companion chatbots and teenagers.

4. Misinformation absorbed as fact

Confident, fluent, wrong answers are absorbed uncritically — especially by pupils who lack the knowledge to spot errors, which is exactly when they're using it.

5. Dependency and confidence erosion

A pupil who cannot start work without AI has lost something. This creeps up gradually and is easily missed.

6. Unequal access

Pupils with paid tools and supportive parents gain; pupils without do not. An equity issue with safeguarding overtones.

The one to prioritise: a child in distress confiding in a chatbot rather than an adult. Everything else on this list is manageable through teaching and rules — this one is about a young person's route to help being quietly diverted, and it deserves specific attention in your pastoral thinking.

What schools should do

Keep it proportionate

Most pupils using AI for homework come to no harm, and the answer isn't alarm. It's the same posture schools take to the rest of online life: teach it, be alert to the vulnerable few, and keep talking to parents. See should my child use AI for homework.

Want your staff and parents briefed properly?

We cover the AI risks that don't make the headlines but matter most. AskColin gives schools low-cost monthly support with one-to-one help whenever you need it — so when you're stuck on a real job at 8am, there's someone to ask. Practical, jargon-free, built around your team.

Get one-to-one AI support for your school

Frequently asked questions

What are the safeguarding risks of pupils using AI for homework?

Exposure to harmful or inaccurate content without school filtering, disclosure of personal data to commercial third parties, disclosing distress to a bot that can't safeguard them, absorbing misinformation as fact, dependency eroding confidence, and unequal access between pupils.

What's the most serious AI homework safeguarding risk?

A child in distress confiding in a chatbot rather than an adult. Other risks are manageable through teaching and rules, but this one quietly diverts a young person's route to help and deserves specific attention in pastoral thinking.

Do pupils share personal information with AI?

Routinely — name, school, address, family details — without any sense it's going to a commercial third party. Younger pupils in particular treat a chatbot as private, which is why explicit teaching about what not to share matters.

How should schools respond to pupils using AI for homework?

Teach explicitly what not to share and why AI can be wrong, tell pupils clearly that AI can't help in a crisis and remind them who can, design homework thoughtfully, brief pastoral staff, reflect it in policy, and tell parents who mostly don't know these risks exist.

← All articles Get AI support for your school →