A parent watching their child use an AI assistant for homework may ask, “is this cheating?” That cannot be answered from the tool alone. The same action — asking a chatbot to explain a concept, check an answer, or draft a paragraph — may be permitted, required, restricted, or prohibited depending on the assignment, institution, age, product, and approved accommodations. A separate learning question is whether the use supports the skill being taught or performs it for the learner.
This article provides a four-question screen for a conversation with the learner and responsible educator. It cannot classify misconduct on its own. The current assignment and institution rules control academic expectations; an approved accommodation, accessibility plan, or official exception must be considered through the institution’s process rather than guessed at home.
School rules control, not household preference
Before using the screen, locate the current instructions for the school, course, teacher, and assignment. Also check any approved accommodation or accessibility plan and the product’s age and account requirements. A household preference does not override an assignment restriction, and it should not remove an institution-approved support.
Check the syllabus and assignment instructions. If they are silent, inconsistent, or unclear, ask the responsible educator before submitting work. Record the answer when the consequences matter. Do not infer permission from silence.
Do not help a learner conceal AI use when disclosure is expected or required. Preserve drafts and source notes where the institution asks for process evidence. If a possible violation has already occurred, use the school’s fact-finding or appeal process rather than asking a chatbot to decide guilt or draft a false account.
The four-question decision tree
For any specific homework task, ask these four questions in order.
1. What is this assignment actually measuring? Not the subject — the skill. A math worksheet on long division is measuring whether the child can execute the algorithm, not whether they can produce a correct final number. An essay assigned after a unit on argument structure is measuring whether the child can structure an argument, not whether a document exists with the right conclusion. Name the skill in one sentence before doing anything else.
2. Does the intended use replace that skill or support it? “Explain long division to me differently than my teacher did” supports the skill — the child still has to execute it afterward. “Solve these ten problems for me” replaces it entirely. “Help me find three counter-arguments to my thesis” supports argument-structure practice; “write my thesis paragraph” replaces it.
3. Does the assignment require disclosure or process evidence? Some educators require students to describe how AI was used, cite it in a specified format, preserve drafts, or use only approved tools. Follow the exact instruction. Omitting a required disclosure can itself breach policy, but only the institution’s process can determine the consequence. See the family agreement’s disclosure norm in the family AI agreement for a household conversation prompt, not a substitute for school policy.
4. What evidence would show the learner understands the skill? If a learner used AI to help draft an essay, can they explain the structure in their own words? If AI helped with a math concept, can they attempt a comparable problem using the supports normally allowed? This is a learning check, not a diagnostic test or automatic misconduct finding. A learner may need more teaching, a different explanation, or an approved accommodation; inability to demonstrate a skill once does not prove why.
Applying the tree: three worked examples
Example 1 — vocabulary practice. Skill measured: recall and usage of assigned words. Proposed use: asking AI to generate five example sentences, which the learner then studies. Replace or support? It may support practice, but the assignment rule still decides whether the tool and generated examples are allowed. Evidence of learning could be using the word in a new sentence under the supports normally permitted.
Example 2 — a history essay with a stated thesis. Skill measured: constructing an argument from evidence. Proposed use: asking AI to write the full essay from the learner’s outline. That use performs much of the apparent writing and argument construction. If the assignment prohibits it or requires original authorship and the learner submits the text as their own, it may breach the policy. If AI use is explicitly required, the educator should define what the learner must author, disclose, and demonstrate. See make AI disagree with you for a possible pressure-test after the learner has produced their own argument, if the assignment permits it.
Example 3 — a coding assignment with a debugging component. Skill measured: locating and fixing a specific class of bug. Proposed use: pasting the broken code and asking AI to identify the fault, then reading the explanation before fixing it. Whether this is permitted depends on the assignment. A comparable unaided or normally supported debugging task may give the educator useful evidence of learning, but it does not override a prohibition or determine misconduct by itself.
Verification habits that transfer either way
Whether or not AI use is permitted, teach the source-checking habit in a child’s first AI conversation: generated text can be fluent and wrong, so check important claims against the assigned or authoritative source. Detection is not a substitute for fair evidence gathering. OpenAI’s current educator guidance says its detector research was not reliable enough for judgments with lasting consequences and notes false positives. Research published in Patterns found higher false-positive rates for non-native English writing in the detectors it tested (Liang et al., 2023); do not generalize that study into a fixed rate for every current detector or learner.
If a school uses an AI-detection tool, ask how it is validated, how other evidence is considered, and how a learner can respond to or appeal a flag. Do not use a detector score alone to accuse or punish a child. Continue teaching the skill and documenting permitted supports regardless of the score.
The escalation rule
If a specific use is unclear, ask the responsible educator with a concrete example: “For this assignment, may the learner use AI to check grammar after writing the draft, and how should that use be disclosed?” Use the school’s accommodation or accessibility contact when the tool is part of an approved support. For an existing allegation, follow the institution’s evidence and appeal process.
Try it today
Pick one low-stakes homework task. Locate the current rule, name the learning objective, record the proposed AI use and disclosure, and ask what evidence of understanding the educator expects. Invite the learner into the conversation in an age-appropriate way. If understanding is unclear, treat that as a reason to ask what teaching or approved support is needed, not as proof of dishonesty.
The homework AI-use decision tree is a printable discussion aid. It does not override assignment rules, approved accommodations, age requirements, or the institution’s misconduct process.
The review also used UNESCO’s guidance for generative AI in education and research and its AI competency framework for students as human-centred, age-appropriate policy anchors. These sources do not approve a specific tool, assignment use, or household decision.
