A19

Expertise self-declaration

Tell the AI what you already know up front so support starts on the right rung, instead of enduring a beginner lecture you didn't need, or a specialist fog you weren't ready for.

When to use this: Before you start studying
Learning impact: Unlocks better later work
Learning evidence: Credible with caveats
Where you are with the topic: Meeting the idea
How often to use it: Useful once for a topic

When to use it

You are about to start a new unit and want a plan before you open the textbook.
You keep meaning to revise, but nothing on your calendar says when.

What to ask AI

Type something like: “Before you teach anything, calibrate to me:”.

Tell it your current level and ask it to stay one step harder than comfortable.

Keep the hard part on your side: you retrieve, attempt, explain, or decide before you ask for help.

How it helps

Default AI tutoring aims at a vague middle.

If you're below it, you nod and drown.

If you're above it, you get analogies and pep and none of the edge you came for.

Either way you waste the session calibrating by frustration.

Expertise self-declaration is the opening move that prevents that: state prior knowledge, goals, and off-limits help styles before content starts. It is not bragging and it is not a diagnostic test (that's Diagnostic pre-test / Prerequisite diagnosis).

The mechanism is simply contingency (same family as Contingent hinting): help matched to state.

Here the state is declared rather than inferred from errors, faster at session start, weaker as a truth serum (people mis-estimate), so pair with a quick probe when stakes are high.

How to run it

Next: before the first explanation, declare: what you've already done, what fails, what you want today, what not to do.
Ask the AI to restate your level in one line and propose a starting rung.
Next: correct the restatement. This catches flattery and underestimation.
Next: begin. If the first answer ignores your declaration, paste it again, models drift.
Update mid-session when you discover you were wrong about your level ("I said I knew eigenvectors; I don't").

Example exchange

You

Before you teach me recursion, calibrate to me: I've written loops confidently, I've seen recursive factorial once and bounced off the base case, and I confuse stack frames with "the function running twice."

Given that, I won't restart from what a function is. What's the smallest example where you can already predict the base-case return value?

You

factorial of 1 should return 1. factorial of 0 I'm unsure about.

Then we start at the base-case boundary, not at loops. Write what factorial of 0 should return and why, before any tutorial.

Copyable prompt

Before you teach anything, calibrate to me:
- I have already: [ ]
- I can do: [ ]
- I consistently fail at: [ ]
- Today's goal: [ ]
- Do NOT: [e.g. start from definitions / give full solutions / use water analogies]
Restate my starting rung in one sentence and wait for confirmation.

The Tell

Here is how you know this method has flipped: twenty minutes in, you're still negotiating pitch. Declaration failed or wasn't enforced.

In that moment the AI (or the schedule, or the story) did the thinking, and you only recognised a finished product.

Tighten the prompt for expertise self-declaration so you attempt, decide, or retrieve before anything is handed to you.

Principle evidence

Strength of the underlying learning idea, not a claim about AI products.

The underlying learning idea is rated moderate. Self-declared prior knowledge is a practical intake for adaptive instruction and expertise-reversal awareness (Kalyuga / CLT): support level should track expertise. Self-report is noisy versus dynamic diagnosis (Prerequisite diagnosis). Principle moderate as an enabler; standalone learning effect is weak, impact is mostly calibration hygiene.

AI delivery evidence

Whether an AI tutor delivers this method well is a separate question.

The learning principle may be moderate, but that does not prove an AI session delivers it well. Treat AI delivery as speculative unless a study of tutoring with this method is named, and none is claimed here.