This is how to never accept AI’s first draft

This is how to never accept AI’s first draft.

Here’s a quiet fact about how these systems work: the first answer is, almost by design, the most statistically average response to what you asked. Average is fine when you’re asking for a packing list. It’s a problem when the thing matters, and the fix isn’t a better first prompt. It’s refusing to stop at turn one.

Most people use AI like a vending machine: one coin, one snack, walk away. It’s a conversation partner. The entire skill is staying at the table for two more turns.

Watch a real email get good

The situation: asking a landlord to fix a heater that’s been broken for two weeks. Turn one:

Turn 1, the ask:

Write an email to my landlord about the heater that’s been broken for two weeks.

Turn 1, the draft:

Dear [Landlord], I hope this finds you well. I’m writing to follow up regarding the heating unit, which has unfortunately been non-operational for approximately two weeks. I would greatly appreciate it if this could be addressed at your earliest convenience...

Grammatical. Spineless. It sounds like nobody, because it’s the average of everybody. Now the moves:

Turn 2, push back:

Too polite. It’s been two weeks and it’s 52 degrees inside at night. I want firm but not hostile, I want a specific deadline, and mention that I’ve texted twice with no reply.

Turn 3, make it check your blind spot:

What am I entitled to here that I didn’t ask about? Ask me where I live if that matters.

Turn three is where it mentioned habitability requirements, asked the state, and suggested documenting the indoor temperature, which is the sentence that ended up mattering. The final email got a repair scheduled in two days. The first draft would have gotten a "thanks, we’ll look into it."

The four follow-up moves

You don’t need many. These four cover most situations, and they work on anything, not just writing:

1. "What would you push back on?" Forces it out of agreeable mode and into review mode. Works eerily well on its own output.
2. "What did you assume?" First drafts are built on invisible guesses about your situation. Making the guesses visible lets you correct the two that are wrong.
3. "Make it sound like me, not like a form letter." Then paste two sentences you actually wrote, from any old email, so it can hear you.
4. "What am I not asking that I should be?" The blind-spot move. This is the one that surfaces the habitability law, the warranty clause, the question for the doctor.

When to stop

Iteration has a failure mode: turn seven, where you’re polishing a sentence nobody will read twice. The signal to stop is when a turn changes style instead of substance. Two to four turns captures nearly all the value. Past that, you’re procrastinating with extra steps.

And a limit worth naming: iteration sharpens what’s there; it doesn’t fact-check it. A confidently wrong claim survives four rounds of polish. Verification is a separate skill in this series, and no amount of "make it better" replaces it.

The habit

For one week, never use the first draft of anything that matters. Ask "what would you push back on?" before you copy a single word. The gap between what you were about to send and what you send instead is this skill, made visible.

This is how to use AI.