Our position on AI
Parents ask us a version of the same question before anything else: is this cheating? Here is our answer in plain terms — what we teach, what we refuse to do, and where we will tell you we do not know.
Last reviewed August 2026
We teach students to use AI the way a working engineer uses a compiler or a careful writer uses an editor: to move faster on work they still own and still understand. AI drafts nothing a student submits as their own. Every AI answer gets checked against a real source before it becomes a study note. A parent or guardian approves anything a minor signs into. We make no promises about a score or an admissions decision, and we say so plainly when the honest answer is that nobody knows.
Six commitments. Each one is specific enough that you could catch us breaking it.
AI can explain a missed problem, quiz a student on a weak skill, or reorganize messy notes. It does not produce work the student turns in as their own. If a workflow only works when the student stops thinking, we do not teach it.
We teach the check, not just the prompt. An AI explanation is confirmed against the textbook, the scoring guide, or the original source before it becomes something a student studies from. Students should leave more skeptical of a confident wrong answer than when they started.
We show the instruction that produced each output. A student who can read and edit their own prompt owns the system and can repair it. A student handed a finished bot owns nothing and learns nothing when it breaks.
A parent or guardian approves the tools, the accounts, and the consent for anything a minor uses. We do not ask a student to open an account that their age or a provider’s current terms do not permit.
Student work is not sold and is not handed to a model provider to train on. We build from the family’s own notes and from materials the family is authorized to use — never another tutor’s or test maker’s proprietary question bank.
We make no promises about a score or an admissions decision, because nobody honestly can. Where our evidence is thin, we label it thin rather than rounding it up into a marketing claim.
The other half of a position
Anyone can publish values. These are the specific things we turn down, including when a family asks for them.
“The reason to be strict about this is not compliance. A student who lets AI think for them is buying a worse education at a higher price. The workflow only pays off if the thinking stays with the student.”
Using AI to produce work a student submits as their own is cheating, and most schools now say so explicitly in their academic integrity policy. Using AI to explain a missed problem, generate extra practice, or quiz a student on a weak skill is studying. The line we teach is whether the student can still do — and explain — the work afterward. If they cannot, the tool was used wrong.
They keep a written error log of missed problems with the reasoning and the correction; they turn feedback from a teacher or tutor into a specific next practice block; and they verify AI explanations against a real source before studying from them. The habit is the product. The tool is replaceable.
We collect what we need to answer your inquiry and run the engagement — contact details, grade, school, and what you tell us in your message. We do not ask for test documents, scores, essays, or accommodation records on an inquiry form. Student work is not sold, and it is not submitted to a model provider for training.
No. We do not promise a score, a score increase, or an admissions decision, and we would not trust a provider who did. What we commit to is a study system the student can run without us, and honest reporting on whether they are actually running it.
Yes, and we would rather they did if it is working. Our test-prep study system is designed to sit alongside whatever instruction a family already trusts — a private tutor, a class, a bootcamp, or a self-study plan. We work on the handoff between sessions, not on replacing the teaching.
It will. Current models state wrong answers with the same confidence as right ones, which is exactly why verification is a taught step rather than a disclaimer. A student who has caught a model being confidently wrong a few times treats every later answer correctly — as a draft to check, not an authority to trust.
Ask it directly. A person reads every inquiry, and “we do not know yet” is an answer we are willing to give.