I read something from Chris Do the other day that stopped me mid-scroll.
He told a story about working with a friend, an AI wizard, years deep in platforms and prompts. They were making YouTube thumbnails and the guy hit a wall. 47 attempts. Each one close but not right. Weird lighting. Strange expressions. That uncanny valley where you can’t explain what’s wrong, but you know something is.
So Chris jumped into Photoshop. Old school. Fixed the face. Adjusted the lighting. Combined three AI outputs into one image. Fifteen minutes. Done.
His friend stared at the screen and said the quietest, most important thing I’ve heard all year.
“I forgot I could do that.”
That hit me. Not because it’s a design problem. Because I see the exact same thing happening in business every single day.
And it should terrify every business owner reading this.
It’s not just happening with thumbnails. It’s happening with strategy. With sales. With marketing. With leadership decisions. Smart people are outsourcing their judgment to a machine and wondering why the output feels flat.
I see it constantly. Business owners who used to trust their gut, who built companies on instinct and hard-won experience, now asking ChatGPT to write their strategy. Letting AI draft their proposals. Getting a language model to decide their pricing.
And the output is fine. Technically correct. Grammatically sound. Reads well enough.
But it doesn’t feel like anything. It doesn’t carry 20 years of knowing when a client is about to walk. It doesn’t have the scar tissue from that deal that went sideways in 2019. It can’t smell a bad partnership from three emails in.
You can.
Or at least, you could. Before you started one-shotting everything.
The one-shot trap
Here’s what happens. You use AI for something and it works. Brilliant. So you use it again. And again. Each time, you hand over a little more. First it’s the admin. Then the first draft. Then the thinking behind the first draft. Then the decision about what to draft in the first place.
Before you know it, you’re not using AI as a tool. You’re using it as a replacement for the thing that made you good.
Your judgment. Your taste. Your experience.
And the worst part? You don’t notice it’s gone until someone reminds you.
The 10-80-10 rule
I think about this as three layers.
The first 10% is you. The intent. The direction. The “what are we actually trying to do here and why.” That’s strategy. That’s leadership. That’s the bit AI genuinely cannot do because it has no skin in the game.
The middle 80% is AI. The heavy lifting. The repetitive labour. The drafting, researching, reformatting, scheduling, number crunching. Give it all of that. That’s what it’s built for.
The final 10% is you again. The refinement. The judgment call. The “this is almost right but the tone is off” or “this misses the point entirely, start again.” That’s your craft.
Most people have quietly expanded that middle 80% to about 95%. They’ve let the first and last 10% atrophy. And they’re wondering why everything feels generic.
The leadership problem
This isn’t just a productivity issue. It’s a leadership issue.
If you’re running a business and you’ve stopped applying judgment to your own strategy, you’re not leading. You’re delegating to a machine that has no understanding of your market, your team, your values, or your customers.
AI is brilliant at the how. Always has been. But it’s useless at the why. And the why is the only thing that separates your business from every other business using the same tools.
Pick your skills back up
The businesses that win from here aren’t the ones using the most AI. They’re the ones that know exactly where AI stops and human starts.
So ask yourself three questions.
- What decisions have you stopped making yourself?
- What skills have you let go soft?
- What happens if you pick them back up?
AI didn’t replace your creativity. It just made you forget you had it.
Time to remember.
PS: I use AI every single day. It saves me 3-4 hours of work. But the strategy, the voice, the decisions about what to say and who to say it to? Those are mine. The 10-80-10 split isn’t a framework I teach. It’s how I actually work. The moment that middle section creeps past 80%, I know I’ve stopped doing my job.
