AI Won't Decide for You. It'll Get You Unstuck.

There's a decision on your desk right now that's been there too long. Maybe it's whether to hire, whether to raise your prices, whether to drop a client who's more trouble than they're worth, whether to sign the lease. You've thought about it in the shower. You've almost brought it up three times. And it's still sitting there.

Here's the thing about decisions like that. They rarely stall because the answer is hard to find. They stall because your thinking is going in circles, and you can't see the shape of your own reasoning from the inside. You keep landing on the same three worries and the same two hopes, in the same loop, and the loop feels like progress.

That loop is exactly what AI is good at breaking. Not by deciding for you, and not by being smarter than you about your own business. It gets you unstuck by doing something a good friend with a legal pad would do: laying the decision out flat so you can actually look at it.

Let me walk through how to use it that way.

Why "should I do X?" is the wrong question

Most people, when they finally turn to AI for a decision, type some version of "should I do X?" and wait for a verdict.

Don't do that. For one, the answer is worthless, because the AI doesn't know your business, your risk tolerance, or the stuff you didn't put in the message. For another, a verdict is the last thing you need. You already have plenty of opinions floating around, including your own. What you're missing is structure.

So instead of asking it to decide, ask it to organize your thinking and then poke holes in it. That's a completely different job, and it's one AI is genuinely useful for.

A four-step way to think it through

Here's a sequence you can run in about fifteen minutes. Do it in one conversation so the AI keeps the context as you go.

Step 1: Dump everything, unfiltered

Open a chat and just brain-dump the decision. Don't organize it. Type it the way it actually lives in your head: "I'm trying to decide whether to hire a second person. Business is up but not steady. I'm exhausted. Cash is okay but not great. I'm worried I'll hire and then things slow down. But I'm also turning away work."

Then ask:

That last part matters. A lot of stuck decisions are stuck because fear and logic are tangled together, and pulling them apart is half the battle. Seeing "exhausted" sorted into one column and "cash flow is uneven" into another is often the first clear breath you've taken about it in weeks.

Step 2: Ask what you're not seeing

Now you go looking for your own blind spot.

This is where AI earns its keep. It doesn't have your emotional attachment to the situation, so it'll surface the boring practical stuff you've been flinching away from. Have you priced out what a slow quarter actually looks like? What happens to the decision if your biggest client leaves? Is there a version of this that isn't all-or-nothing?

You won't agree with all of it. That's fine. You're not looking for answers here, you're looking for questions you forgot to ask.

Step 3: Make it argue against you

This is the step people skip, and it's the most valuable one.

Whatever you're leaning toward, tell the AI to argue the other side as hard as it can.

Then flip it: "Okay, now argue just as hard for hiring now."

Reading a strong version of the argument against your gut does one of two things. Either it surfaces a risk you'd been quietly ignoring, and you're glad you caught it before signing. Or it doesn't land, and you notice that none of the counterarguments actually move you, which is its own kind of answer. Both outcomes get you unstuck. Being able to argue both sides is one of the things that separates a good decision from a hopeful one.

Step 4: Turn it into a small next step

Big decisions feel stuck partly because they feel like one giant irreversible leap. Usually they aren't.

Maybe it's a thirty-day contractor instead of a hire. Maybe it's a test price on one segment before you change everything. Maybe it's a single conversation with your accountant. The AI is good at breaking a scary leap into a first step you can actually take this week.

That first step is usually what you were missing. Not the answer, just permission to move an inch.

What this does not do

It's worth being clear-eyed here, because this technique is genuinely useful and also genuinely limited.

It does not know your business better than you. Everything it says is built from what you told it plus general patterns, so a confident-sounding answer can still be wrong for your specific situation. Treat its output as a sharp thinking partner, not an oracle. This is the same posture we've written about in judging work you didn't do: the drafting gets handed off, the evaluating stays yours.

It should not make consequential calls for you, especially anything with legal, financial, or people implications. A layoff, a big loan, a partnership split. Use it to think those through, then bring the actual decision to a human who knows your situation, and to your own judgment. Deciding what to hand off and what to keep is its own skill, and we broke that down in the difference between automating and delegating.

And it will happily agree with you if you let it. If you only ever ask it to support your leaning, it'll cheerfully build you a case for whatever you already wanted. The whole value is in Step 3, the part where you make it push back. Skip that and you've just built yourself a very articulate yes-man.

Why this beats stewing on it

Think about what actually happens when a decision sits on your desk for a month. You don't gain information. You just carry it around, spending a little attention on it every day, letting it drain focus from everything else. The cost of a stuck decision isn't just the delayed outcome. It's the ambient tax it charges on the rest of your week.

Fifteen minutes of structured thinking doesn't guarantee you'll choose right. Nothing does. But it reliably gets you from "circling" to "moving," and moving is where the real information comes from. You learn more from one small reversible step than from another month of turning it over in the shower.

None of this requires you to be technical. It requires you to be willing to look at your own reasoning honestly, which is exactly the kind of broad judgment that's getting more valuable, not less.

Pick the decision that's been sitting there the longest. Open a chat. Dump it out and ask it to argue with you. Worst case, you spend fifteen minutes and confirm what you already thought. Best case, you finally put the thing down.

If you're trying to figure out where AI actually fits into your business or daily workflow, that's exactly what we help people do at Humanity AI.