Give context so the answer is about you

Ask an AI for a training plan and you get the plan for a person with healthy knees, free evenings, and a gym membership, because with nothing about you in the question, the model answers the average person who asks it. This lesson teaches the move that fixes it: loading your question with your situation, your constraints, what you already tried, and what the answer is for, so the answer lands in your life. You'll run it this week on a real question, one you already asked and got a shrug of generic advice for, and you'll watch the second answer do things the first one couldn't. You don't already have this move because the bare question looked like it worked: an answer came back, and it belonged to someone else. The same move, done sharply, is a thing adults get paid for, and you can have it at 17.

Why does a bare question get a generic answer?: With nothing about you in the message, the model answers the question as most people ask it: the average asker's version, assembled from its training data.
What does a constraint (time, money, equipment) do to an answer?: It deletes the branches you can't use before they reach you. "No gym membership" removes gym plans from the answer space.
Why tell the model what you already tried?: The failed option leaves the table, and the way it failed tells the model what the replacement has to survive.
What changes when you state what the output is for?: The target. Deciding, preparing, and negotiating need different answers to the same question, and an unstated purpose gets guessed as the common one.
The test for whether a fact belongs in your ask?: If this fact changed, would the answer have to change? If yes, it goes in. If no, it stays out.
Why leave inert personal details out of the ask?: Including a fact marks it as relevant. The model can't tell "this binds my week" from "this was on my mind," so a decoy can steer the answer.
A fresh chat opens. What does the model know about you?: Close to nothing you can rely on. Memory features save fragments at best; a question that carries its own context works in any chat, any app.
A fact feels too personal to share. What's the move?: Send the constraint without the identity. The budget number changes the answer; your name attached to it changes nothing.
What's the adult version of loading a question well?: Writing a brief: describing an outcome sharply enough that someone else could build it. People are paid for that skill, and it trains on questions like yours.

You ask "is a gap year a good idea?" and get a balanced list of pros and cons that could run in a magazine. Why did the answer come out generic?: With no details in the question, it answers the question as most people ask it, the average asker's version
You want a training plan. Which single addition changes the plan the most?: "I can train Tuesday and Thursday evenings, and my left knee can't take running or jumping yet"
Why include "I tried a five-day gym plan in April and quit in week two" in your ask?: It takes the failed option off the table, and how it failed tells the model what the replacement has to survive
You paste a summer sublet contract into an AI and ask "explain this contract." What changes if you add "I'm deciding whether to sign by Friday, and I've never rented before"?: The answer reorganizes around the decision: what's unusual, what's risky for a first-time tenant, what to ask before Friday
Your friend wants AI help planning a month of food spending but won't give it the budget number: "too personal for a chatbot." What's the real trade-off?: Without the number, the answer gets written for an average budget, and average is what their month can't afford
You give the AI your situation in three long paragraphs, including that your cat has been sick, and the study schedule that comes back keeps reserving time "for veterinary appointments." Why?: You included the cat, so the model treated the cat as relevant, because you're the one who decides what belongs in the message
You've used the same AI app most days for a year. You open a new chat and ask a bare question. How much of your situation does it know?: At most fragments, if a memory feature saved some, and nothing you can rely on being there for this answer
Two summer jobs, and you want the AI's help choosing. Which ask earns its answer?: "Warehouse pays 11.20, weekday days. Café pays 9.80, evenings and weekends. I need at least 1,100 euros by autumn, I have handball camp the first week of July, and I want a reference that helps with sales jobs next year. Which fits, and what am I not weighing?"