Force the answer into a decision matrix
Lesson 6 of 7 · Compliance & safe AI use
1 · Learn the move · Comparison matrix
A wall of prose hides the trade-off; a matrix exposes it. The skill is naming the exact rows and columns so the options land side by side on the same criteria - one row each - instead of three paragraphs you have to transpose in your head. What you get is a table you can defend in lab meeting, with the gaps marked instead of guessed.
Compare these options in one table: one row per option, these columns: [criteria]. Use only the facts I give; mark anything missing UNKNOWN. End with the one-line trade-off. [options]
2 · Your turn. You write the prompt
Three normalisation strategies are on the table and the group keeps arguing in prose. Write a prompt that turns them into one matrix - what each assumes, what it costs in sample, what it does to the variance, how it fails - so the choice is legible.
Write the prompt for this scenario as given, or swap in a real one from your own work. The score works the same either way.
Remember: the AI sees only your prompt, not this page. If the situation isn't in your prompt, it doesn't exist.
Optional. These shape the output when you run your prompt below, not your score.