It's a reasonable question, and it deserves a real answer rather than blanket reassurance or blanket alarm. Here's what matters when you're deciding whether an AI-powered learning tool is appropriate for your child.
The questions worth asking
1. What data does it collect about my child, and why?
A well-designed children's product collects the minimum needed to work. For an exam-prep tool, that's typically a first name, year group, and the practice and progress data your child generates. If a product asks for your child's date of birth, home address, school name, or photo without a clear reason, question it. In the UK, the ICO's Age Appropriate Design Code (the "Children's Code") sets standards for services likely to be used by children, including data minimisation. A reputable UK product should be able to show you how it applies this.
2. Who controls the account, my child or me?
Products built with care for this age group put the parent in control of account creation and settings, with the child working inside that account rather than signing up independently. That is the structural difference between a tool designed for a supervised child and one built generically for "anyone".
3. What happens if my child says something concerning to the AI?
A responsible AI tutor should be explicitly instructed to steer off-topic or distressing input toward a trusted adult, rather than trying to "counsel" a child itself. An AI system is not a substitute for a parent, teacher or appropriate professional in a genuine safeguarding situation. Ask, or check the product's documentation, how this is handled.
4. Is my child's data used for advertising or sold to third parties?
The answer should be a clear no in any product's privacy policy, stated plainly rather than buried in dense legal language. If it isn't clearly stated, ask directly.
5. Where does the data actually go?
Many AI tools rely on a third-party AI provider to generate responses. That is common and not automatically a red flag, but a transparent product will tell you which third parties see your child's input and under what safeguards, including whether the provider uses the data to train its own models. Check the product's privacy policy and the provider's terms rather than assuming.
What "good" looks like in practice
- A parent-controlled account, with the child working under it
- Clear data minimisation, with no unnecessary personal details collected
- An explicit safety instruction built into how the AI responds to off-topic or concerning input
- A working way to flag a concerning response, so a real person can review it
- No advertising, no behavioural profiling, no data sold to third parties
- A published privacy policy that names the actual data categories, retention periods and third parties involved, rather than vague boilerplate
A healthy level of scepticism is appropriate
AI-generated educational content can occasionally be wrong, unclear or inconsistent. That is true of any AI system, not a defect of one product. A responsible product should say so plainly rather than overselling, and should make it easy for you to review your child's sessions and flag anything that looks off. If a product's marketing promises perfect, guaranteed results, treat that as a bigger warning sign than any technical detail. No exam-prep tool, human or AI, can guarantee an exam outcome.
The bottom line
AI tutoring tools aren't inherently safe or unsafe. It depends on how the specific product is built, what it collects and how open it is about both. The five questions above give you a real basis for evaluating any product, including ours. You can read PassEleven's own Privacy Policy and Terms of Service and judge for yourself.
Related guides
This guide is general information for parents evaluating AI-based educational tools.