The moment AI starts agreeing with you too quickly and too easily is the moment to slow down and think first.
Because if AI isn't helping you refine, improve, and stress test your thinking, it's just helping you move. Probably in the wrong direction.
That distinction matters more than most people realise. Over the last few posts, I've talked about false confidence and why AI agrees with everything you say. AI can make smart people confidently wrong, and therefore move faster in the wrong direction. But the real issue isn't AI. It's when people blindly follow it, abdicate responsibility, and quietly diminish their own authority.
Why should you think first before using AI?
For me, AI never sits before the first iteration of thinking. By the time I open up AI, I've already got my own collection of ideas, some data, some facts, a few hypotheses, and a best guess about the direction I think things will go. That thinking has happened already. AI comes in after.
And there's one role I never let it play: AI never decides for me. It can say "if I were you, I would do X because of Y." That can be useful. But the decision itself rests with me.
This sounds obvious. But in practice, the line blurs faster than you'd expect. AI is so fluent, so confident, so quick to present a clear recommendation, that it's easy to treat its output as a decision rather than an input. Especially when you're busy and the answer sounds right. That's not AI being dangerous. That's convenience replacing judgment.
What happens when people hand their thinking to AI?
The real issue isn't bad outputs. It's abdication. When someone uses AI to skip their own thinking, when they want it to do the hard part, the heavy lifting, which is the judgment, they're not thinking with it. They're trying to think less. And that's how your authority quietly gets handed away.
Your authority as a business owner, as a leader, as someone people trust, comes from judgment. Not speed. Not output volume. Not having an answer for everything. Judgment.
When you outsource judgment to a tool, the authority doesn't transfer to the tool. It just disappears. Your clients, your team, your partners trusted your thinking. If that thinking now comes from a machine, the thing they valued is gone. And they'll feel it, even if they can't name it. So the question to keep in front of you isn't "is AI useful?", it clearly is. The question is: am I using it to strengthen my thinking, or to replace it?
How should AI actually fit into your decision-making process?
The best use of AI for me is as a thinking partner. I use it to stress test ideas, to look for blind spots, and to play devil's advocate. I want AI to serve as a board of advisers and challenge me. Not just agree with me. If it's nodding along, it's not adding any value, and there's a good chance it could be cheerleading you off the edge of a cliff.
Here's the practical sequence I use:
- Think first. Form your own position before opening AI. Have ideas, data, hypotheses, even rough ones.
- Stress test second. Ask AI to find the holes, the risks, the second-order consequences. "Tell me why this won't work" before "help me plan this."
- Build third. Only after your thinking has survived challenge do you use AI to develop, expand, and refine.
- Decide always. AI advises. You decide. That order is non-negotiable.
That sequence keeps AI in the right seat. It's contributing to your thinking, not substituting for it.
How do you know a decision is genuinely yours?
I know a decision is mine when it sits right in my gut and in my head. If I'm willing to own the upside and the downside, without outsourcing responsibility or my thinking to a robot, that's when it feels right. And that's when I commit.
There's a difference between "AI told me this is a good idea and I'm going with it" and "I've thought this through, I've pressure-tested it, and I'm backing it." The first one is delegation. The second one is decision-making. And your clients, your team, your partners can tell the difference. Not because they know you used AI, but because confidence that comes from genuine thinking sounds different from confidence that comes from borrowed certainty.
What's the difference between thinking with AI and thinking less?
"Thinking with AI" means you arrive with your own perspective, your own questions, your own context, and you use the tool to see what you might be missing. The thinking gets better because of the collaboration.
"Thinking less" means you arrive empty-handed and let the tool fill in the blanks. You skip the uncomfortable part, the part where you sit with uncertainty and form a view, and jump straight to an AI-generated recommendation that feels like clarity but isn't yours.
Both look the same from the outside. The person in both scenarios is sitting at their desk, using AI, producing output. But the quality of what comes out, and the authority behind it, couldn't be more different. Used properly, AI won't make the decisions for you. But it can help you see more clearly and decide better. For serious operators, that's exactly where AI should sit.
Frequently asked questions
What does AI decision making actually mean for business owners?
It doesn't mean letting AI make your decisions. It means using AI as a thinking tool within your decision-making process, to stress test ideas, surface blind spots, and challenge assumptions before you commit. The decision itself should always rest with you.
Why should I think first before using AI?
Because AI responds to the direction of your question. If you arrive without your own thinking, AI fills the gap with plausible-sounding output that feels like clarity but isn't grounded in your context or judgment. Thinking first means you bring substance to the conversation, and AI sharpens it rather than replacing it.
How do I know if I'm using AI properly for decisions?
A good test: can you explain the reasoning behind your decision without referencing what AI told you? If your confidence comes from having genuinely thought it through, that's proper use. If it comes from AI recommending it and you going along, that's abdication dressed up as efficiency.
What's the risk of letting AI make decisions for me?
The risk isn't bad answers. It's that you quietly lose your authority as a thinker. Your clients, team and partners trust your judgment. When judgment gets outsourced to a tool, the thing they valued disappears, and the sharpness of your own thinking dulls because you exercise it less.
How is using AI as a thinking partner different from just asking AI for advice?
Asking for advice is passive: you present a problem and wait for a recommendation. Using AI as a thinking partner is active: you arrive with your own ideas and position, then ask AI to challenge, expand or pressure-test them. The first makes you dependent on the tool; the second makes your thinking better.
What's the connection between AI decision making and personal authority?
Your authority is built on judgment. When you outsource judgment to AI, you don't transfer authority to the tool, you lose it. Using AI to support and sharpen your judgment preserves that authority. Using it to skip your own thinking erodes it.
