AI homework helpers vs. cheating: a parent’s guide to school AI policies

The gap between what kids do and what schools say

A July 2026 report in Fortune put the share of students using AI for homework at 84%, roughly double the figure two years earlier. Written school policy has not kept pace.

The gap between the two is a real problem for families. When the rules are vague, kids fill the silence with their own interpretation — and so do teachers. One class may treat AI brainstorming as routine; the next may treat it as a suspension-worthy offence. Your child can be doing the same thing in periods 2 and 4 and face completely different consequences.

The good news: parents can do something practical about this right now, before the next report card lands.

What “cheating with AI” actually means

There is no single definition of AI cheating. The line moves depending on the subject, the age group, and the teacher. But most school policies that do exist draw the line at the same place: whose thinking is in the final work?

Here are examples that fall on each side of that line.

Generally allowed in most schools:

• Using an AI tutor to explain a concept you don’t understand (“Can you explain photosynthesis to me as if I’m 12?”)
• Brainstorming essay topics and then writing the essay yourself
• Asking AI to check your grammar after you’ve written a draft
• Using AI-powered tools built into school platforms (many districts are rolling these out intentionally)

Generally treated as cheating:

• Having AI write the essay, then submitting it unchanged
• Asking AI for the answer to a maths problem instead of solving it
• Paraphrasing AI output to disguise it, then submitting as original work
• Using AI on a test or quiz that prohibits outside tools

Why AI detection tools won’t solve this

You may have heard that schools use software to catch AI-written work. The honest picture is messier. AI-detection tools are unreliable enough that a student’s genuine writing can be flagged as AI-generated — which is why some institutions have stopped relying on them.

Some schools have already faced situations where students were penalised for work they wrote themselves. A false accusation is stressful for a child and difficult for a parent to fight without technical expertise.

This is one reason why conversations and clear policies matter more than detection. Rules your child understands, agreed in advance, are a better protection than a scanner that may not work.

How to find your school’s actual policy

Don’t assume a policy exists or doesn’t exist — ask. Here is where to look and what to ask for.

1. Check the school’s student handbook. Most schools publish this online. Search the PDF for “artificial intelligence,” “AI,” or “academic integrity.” Some districts have updated their Acceptable Use Policy to include AI; many have not.
2. Email the class teacher directly. Ask: “What is your policy on students using AI tools like ChatGPT for homework in your class?” Teachers often have their own rules that differ from school-wide policy. Getting a written reply protects your child.
3. Ask the school office for the district AI policy. Some US states have begun requiring districts to publish an AI policy, Ohio among them. Other states are following. Your district may have published one.
4. Check the back-to-school materials. It may arrive in the school’s welcome letter or the parent portal.

If the school genuinely has no policy, that is worth raising at a parent–teacher meeting or via the parent council. A school without a policy is leaving students in a grey area that can hurt them.

The AI tools worth knowing about

Not all AI study tools are equal. There is a meaningful difference between a tool that hands your child the answer and one that makes them work for it.

Socratic-style tutors (such as Khanmigo from Khan Academy) ask guiding questions instead of giving direct answers. If your child asks “what is the answer to question 3?” it will push back with “what have you tried so far?” These tools tend to reinforce learning rather than replace it.

Direct-answer tools (general chatbots used without boundaries, or homework-solving apps that photograph an equation and return the solution) give the answer immediately. The child gets a correct answer and learns nothing.

The tool itself is not always the problem — how your child uses it is. A general chatbot used to ask “help me understand fractions” is a tutor. The same chatbot used to type in the entire assignment question is a ghost-writer.

For a broader look at how kids are using AI tools day-to-day, see our AI chatbots and kids guide and the AI chatbot safety guide.

Setting ground rules at home

Once you know what the school says, you can build a household rule that matches it. A few questions to settle before the new term:

• Which tools are allowed? Decide together which AI tools your child may use for schoolwork, and which are off-limits for graded work.
• When must they disclose? Some schools now ask students to note when AI was used in their work. Make this the default in your household, even if the school does not require it. It builds the habit of transparency.
• What happens with a no-AI assignment? When a teacher specifically says no AI, that means none — not even for brainstorming or checking grammar. Make sure your child knows this applies even if they could get away with it.

What to do if your child’s school flags their work

If a teacher accuses your child of submitting AI-generated work:

1. Ask for the specific evidence. Schools should be able to say which detection tool was used and what percentage it flagged. That number alone is not proof of anything.
2. Request an oral follow-up. Ask that your child be given the chance to answer questions about their work in person. Genuine understanding — or genuine absence of it — usually becomes clear quickly in conversation.
3. Check your child’s drafts. Most word processors and Google Docs save version history. If your child wrote the work themselves, those timestamps show it.
4. Raise the false-positive issue calmly. Detection tools flag non-native English speakers at far higher rates. Stanford researchers found in 2023 that detectors misidentified over 61% of TOEFL essays as AI-generated while being close to perfect on essays by native-English-speaking US students, because they score writing on “perplexity” — a measure that penalises plainer phrasing. If your child falls into either group, say so.
5. Escalate to the head of year if the teacher will not reconsider and the penalty is significant (e.g. zero grade or suspension).

For more on navigating school-issued devices and what is monitored there, see our school-issued devices guide.