The Maths on AI Is Wrong

Everyone’s talking about AI replacing jobs. The headlines write themselves: millions of roles automated, entire industries disrupted, mass unemployment on the horizon.

But the maths is wrong.

Not completely wrong — AI will replace some jobs. If your work is routine, repeatable, and well-documented, then yes, a model can probably do it faster and cheaper than you can. That’s real, and it’s already happening.

But here’s what the doom-and-gloom predictions miss: if your job is highly skilled or creative, AI isn’t replacing you. It’s actually creating more work.

I started 20-something projects to test this

Since I started using AI seriously, I’ve launched over twenty projects to explore what it’s capable of. Not toy demos — real work. Books, apps, reports, marketing strategies, data analysis.

Here’s what I’ve found: my output is 10–100x what it would otherwise be as a single human.

I can write a 30,000-word book. I can produce a 10,000-word report. I can build apps, create marketing personas, summarise thousands of notes and analyse them for patterns.

That sounds like a productivity miracle. And in some ways, it is.

But there’s a catch nobody’s talking about.

The review bottleneck

All of that output requires review.

Every word of that 30,000-word book needs to be read, checked, and refined. Every insight in that 10,000-word report needs to be validated. Every app needs testing. Every marketing persona needs a sanity check against real-world experience.

And I simply don’t have time.

I’m spending so much time experimenting and generating that I can’t keep up with reviewing what comes out the other side. The bottleneck isn’t production anymore — it’s human judgement.

AI hasn’t eliminated my workload. It’s shifted it. Instead of spending my time creating, I now need to spend my time evaluating. And that evaluation requires exactly the kind of skilled, experienced thinking that AI can’t replace.

The job creation nobody predicted

This is the part the forecasters get wrong. They model AI as a simple substitution: one AI replaces one human, job gone. But that’s not how it works in practice.

What actually happens is:

• Output explodes. One person with AI can produce what used to take a team.
• Review demand explodes with it. Every piece of output needs a skilled human to assess quality, accuracy, and relevance.
• New roles emerge. Someone has to manage the AI workflows, curate the outputs, make the judgement calls, and integrate the results into real strategy.

The more AI produces, the more humans are needed to make sense of it.

Think about it this way: if I can now produce 20x the strategic output I used to, I don’t need fewer people on my team. I need more — people who can review, challenge, refine, and execute on what AI generates.

The skills that matter now

This reframes the entire conversation about which skills are “AI-proof.” It’s not about whether AI can do your task. It’s about whether AI can judge the result.

The skills that become more valuable in an AI-saturated world are:

• Critical evaluation — Can you tell good output from bad?
• Domain expertise — Do you know enough to catch when AI gets it subtly wrong?
• Editorial judgement — Can you shape raw output into something that actually works?
• Strategic thinking — Can you decide which of the twenty possible outputs to actually pursue?
• Taste — Do you know what good looks like?

These aren’t skills that AI eliminates. They’re skills that AI makes more important, because there’s simply more output to apply them to.

The real risk isn’t job loss — it’s drowning

Here’s my honest experience: I’m not worried about AI taking my job. I’m worried about drowning in AI output that I can’t review fast enough.

Every project I start creates a review backlog. Every experiment generates artefacts that need human eyes. The more capable AI gets, the wider the gap between what it can produce and what I can meaningfully evaluate.

That gap? That’s where the new jobs live.

Companies that figure this out early — that invest in review capacity, quality frameworks, and human-in-the-loop processes — will outperform those still debating whether AI will replace their workforce.

The maths, corrected

The old equation: AI replaces human → fewer jobs.

The real equation: AI multiplies output → more output needs review → more demand for skilled humans.

AI isn’t making skilled people redundant. It’s making them insufficient. There aren’t enough experienced professionals to review what AI can now produce. And that’s not a job-loss story. That’s a hiring story.

The maths on AI is wrong. Not because AI isn’t powerful — it is. But because the models assume output is the end of the process. It’s not. It’s the beginning. And everything after output — the review, the refinement, the judgement — that’s still fundamentally human work.

We don’t need fewer people. We need more.

We just need them doing different things.