Why "is AI cheating?" is the wrong first question
Most parents land on this page having already asked the wrong question. "Is AI cheating?" treats AI like a single object with one moral status — like asking "is a calculator cheating?" without specifying whether it's being used on a long-division worksheet or a calculus exam where mental arithmetic is the point. The answer changes completely depending on what's being asked of the tool and what the assignment is actually testing.
A more useful question is: what is my kid asking the AI to do, and does that match what the teacher wants them to practice? Asking a study bot to explain why a step in a proof works, after getting stuck for ten minutes, is functionally identical to asking a parent or a tutor the same question — nobody calls that cheating. Asking the same tool to write a full essay and turning it in as original work is a different act, and it stays a different act no matter how good the explanation-generation happens to be. The tool is the same in both cases. The difference is entirely in the ask and the intent, which is why a blanket rule ("no AI, ever" or "AI is fine, whatever") misses the actual decision parents need to make.
This distinction matters more as these tools get better at explaining rather than just answering. A tool that shows numbered, worked steps — the kind @vustStudyBot generates for a math or science problem — genuinely resembles a tutoring session more than an answer key. But "resembles a tutoring session" is a description of the software's behavior, not a guarantee about how any specific kid will use it in any specific moment. That gap is exactly what a family agreement is for.
The 5-rule starting family AI agreement
This is a starting point, not a finished policy — adjust the specifics for your kid's age and your family's actual homework patterns. The goal is five rules simple enough that a 10-year-old and a 16-year-old can both explain them back to you in one sentence each.
-
AI explains, you write. If a tool explains a concept or shows the steps of a worked example, that's fine to read and learn from. Copying its output word-for-word into an assignment that's graded on your own writing or your own work is not — that's true whether the source is an AI tool, a classmate's paper, or a website.
-
Show me the "next one." After using AI help on a problem, be ready to solve a similar problem — same type, different numbers — without looking at the explanation. This is the single best real-world check for whether the help turned into understanding or turned into copying, and it takes two minutes.
-
Name the tool, out loud, before you use it. "I'm going to ask StudyBot to explain the derivative rule" is a five-second sentence that keeps AI use in the open rather than something to hide. Secrecy is usually the actual red flag, not the tool itself.
-
Some assignments are AI-off, on purpose. In-class tests, timed quizzes, and anything the teacher has explicitly said should be done independently are AI-off, full stop — no explaining exception, because the whole point of those assessments is to measure unaided understanding. Homework and take-home practice problems are usually where explanation-style help is appropriate; check the syllabus or ask the teacher if it's genuinely unclear.
-
We revisit this every few weeks. Homework changes as the school year goes on — a rule that made sense for third-grade multiplication tables won't fit ninth-grade algebra. A short, low-stakes check-in ("is this still working, has anything changed") keeps the agreement relevant instead of becoming a rule nobody remembers agreeing to.
Two conversation scripts you can actually use
Script 1 — Starting the conversation before a problem comes up. This works best proactively, not as a reaction to catching something.
Parent: "I want to talk about the homework-help apps and AI bots you might see other kids using. Have you used any of them, or thought about it?" Kid: [Answer varies — be ready for either "no" or "yeah, everyone uses one for math."] Parent: "Here's the thing — using one to get something explained is totally fine, that's what a tutor does too. Using one to just get the answer and write it down without understanding it isn't really helping you, even if the teacher never finds out. Can we agree that if you use one, you'll be able to show me the next similar problem without looking?" Kid: [Ideally, agreement — if not, ask what feels unfair about it and adjust rather than issuing an ultimatum.]
Script 2 — After you notice AI-generated-looking work. This works best as curiosity, not an accusation — accusing a kid who was actually being honest damages trust for no reason.
Parent: "This explanation in your math homework looks like it might be from an AI tool — can you walk me through this step in your own words?" Kid: [If they can explain it — great, that confirms understanding regardless of the source.] Kid: [If they can't explain it — "It looks like you got help understanding this, but haven't quite gotten there yet. Let's do one more together so it sticks before the test."] Parent (either way): "I'm not upset that you used a tool — I want to know it actually helped you learn the thing, not just finished the homework."
The through-line in both scripts: the goal is a conversation about understanding, not a gotcha about tool use. Kids who feel like admitting to AI use triggers punishment will just hide it better, which is worse for everyone than an open conversation with a clear rule.
Green-flags vs red-flags: judging any study AI tool
Use this table for any AI homework tool your kid uses — not just StudyBot. The signals are about the SHAPE of the help, not the brand.
| Signal | Green flag | Red flag |
|---|---|---|
| What comes back | Numbered, named steps or an explanation of the method | A single final answer with no working shown |
| Follow-up capability | Kid can ask "why does step 3 work?" and get a real answer | No way to ask about a specific part of the reply |
| Self-testing | Tool offers practice questions or quizzes on the same topic | Tool only ever answers the exact question asked |
| Transparency at home | Kid mentions using it, unprompted, like mentioning a tutor | Kid closes the app or chat when you walk by |
| Copy behavior | Kid can redo a similar problem without the tool afterward | Kid can only redo the EXACT same problem, not a variant |
| Assignment type | Practice problems, homework, studying for a test | Timed quizzes, in-class tests, take-home exams marked independent |
| Account/data | Runs through an existing account you already trust (e.g., a messaging app) | Requires a new account with personal details you can't review |
None of these signals is proof on its own — a kid who's transparent about tool use but still copying the output is possible, and a kid who's quiet about it might just be private, not deceptive. Use the table as a set of things to notice, not a verdict from a single data point.
What @vustStudyBot actually does, precisely
To be exact rather than reassuring: @vustStudyBot is a Telegram-based study assistant. Sent a text or photo homework question, it classifies the request (a step-by-step "task solve" versus a "summary/explain" for teaching-style content) and returns structured, notebook-style output — numbered steps for problems, section-style breakdowns for concepts — plus optional self-test quiz questions generated on the same topic. It supports K–12-level routing and multiple curricula.
What it does NOT have: a parental-control mode, an age verification gate, screen-time limits, or a setting that hides the final answer from a solved problem while still showing the steps. If a task solve is requested, the reply eventually reaches the answer — structured as a worked method, not withheld behind a toggle. The "learning-first" framing on this page describes how the output is SHAPED (steps and method, not a bare number) and what quiz/self-test content is available — it is not a claim that the software enforces a house rule for you. That enforcement is what this guide's five rules and two scripts are for.
Where this fits with school policy
Nothing here overrides your child's actual school or teacher's AI policy, which varies widely — some schools ban AI tools entirely for graded work, some allow "AI as a study aid" but not for submissions, and some are actively teaching AI literacy as a skill. When your family agreement and the school's policy might conflict (for instance, your kid uses AI to study for a test the school considers should be prepared for independently, which is fine, versus using it to write graded homework the school explicitly bans AI for), the school's policy on that specific assignment type wins. Treat this guide as the layer that governs everything the school policy doesn't explicitly cover — everyday practice, understanding a concept, and preparing for something the school hasn't given AI-specific instructions about.