By Chuck Gallagher — Business Ethics Keynote Speaker and Trainer
TL;DR: Chuck Gallagher, AI ethics speaker and author, argues that for small businesses the real AI danger isn’t lost jobs — it’s the freed-up owner who now ships three times the work with the same single set of eyes checking it.
A woman I’ll call Diane runs a marketing shop. Six people. She used to spend her mornings writing first drafts — proposals, client emails, the monthly report nobody wanted to write. Then she started using AI. Now those drafts take twenty minutes instead of three hours.
She told me it gave her back her mornings. She wasn’t wrong. She was also missing something.
Here’s the thing she missed. The drafts still go out under her name. The client still reads them as her judgment. And the only person checking whether the AI got it right — the only person who ever checked — is still her. Except now there’s three times as much going out the door. That’s the part nobody puts on the brochure.
There’s a piece making the rounds in business circles right now, out of Harvard Business Review, arguing that AI is crushing middle managers. The logic holds for big companies. Work speeds up at the bottom, ambition expands at the top, and the people in the middle get buried catching the errors. Gartner predicts that in 2026, one in five organizations will use AI to flatten their structure and cut more than half their middle-management roles. For a large firm, that’s the risk. Subtract the layer, lose the people who catch what breaks.
But most of America doesn’t work at a large firm. Most of America works at Diane’s shop. And Diane doesn’t have a middle layer to cut. She is the middle layer. She’s also the top layer and, on a bad week, the bottom one too.
As an AI ethics speaker and author, I spend a lot of time in rooms full of owners like Diane — not Fortune 500 boardrooms, but the companies where one person does four jobs and just handed a fifth to a machine. What I tell them is this: the danger at your scale isn’t the danger you’ve been reading about.
What’s Actually Happening Inside Small Businesses?
The numbers tell a quieter story than the headlines. Roughly 68% of small businesses now use AI regularly. But about 77% of them have no written policy governing it — no standard for what counts as good enough, no rule about what an AI tool is allowed to touch in client-facing work. Two-thirds are running the engine. Three-quarters never wrote down how it’s supposed to run.
And here’s what makes it different from the big-company story. About 62% of small business owners use AI themselves, hands on the keyboard. They’ve watched it invent a statistic. They’ve caught it making up a citation. So they keep a human hand on the things that matter — the finances, the insurance, the client relationship. That instinct is good. That instinct is also the whole defense. There’s no second set of eyes behind it. There’s just the owner’s gut, and the owner’s gut gets tired.
So Is AI Taking the Jobs?
No — and that’s exactly what hides the risk. The layoff story is an enterprise story. On Main Street, only about 8% of owners report cutting a role because of AI, and nearly 70% haven’t cut anyone at all. One survey of small employers found 98% reported no change in headcount. Another found 82% actually grew their workforce. Nobody’s getting buried by a layoff here.
What’s happening instead is subtler. AI takes over the tasks the owner used to do by hand — the late-night emails, the first drafts, the data entry — and gives the owner back time. Five hours a week, by one count. Sounds like a gift. But look at what fills those five hours. More proposals. More client work. More going out the door. The throughput goes up. The number of people checking it stays at one.
That’s not relief. That’s a risk transfer. And nobody signed off on it.
Why Does This Matter More Than the Headlines Suggest?
Because the consequence is delayed and it’s quiet, which is precisely the kind of consequence people talk themselves out of seeing. I learned that the hard way once, in a season of my life I’ve spent the years since trying to make useful to other people. The choices that ruin you rarely announce themselves. They feel like efficiency. They feel like getting ahead. The bill comes later.
When I work with a company as an AI ethics speaker and author, the first question I ask isn’t which tools they’re using. It’s: who reads the work before the client does? In a big firm, the honest answer used to be a manager. In a small firm, the honest answer is nobody — and AI just tripled how much there is to not read.
A consulting firm that hollows out its review process finds out in five years, when its pipeline runs dry. A six-person shop finds out faster. One unchecked deliverable, sent to the client who happens to be 30% of revenue, with a number in it the AI made up. That’s not a slow erosion. That’s a Friday.
Every choice has a consequence. Adopting AI is a choice. Adopting it without deciding who checks the work is also a choice — just a quieter one. The owners who come through this well won’t be the ones with the best tools. They’ll be the ones who treated their own freed-up time as something to spend carefully, not something to fill.
Frequently Asked Questions
Does AI cause layoffs at small businesses?
Mostly no. Survey data shows only about 8% of small business owners cut a role because of AI, and the large majority report no change in headcount — some even grew. At small scale, AI tends to absorb tasks the owner was doing personally rather than replace an employee.
What is the biggest AI risk for a small business?
The lack of a checkpoint. Around 77% of small businesses using AI have no written policy, which means no agreed standard for accuracy, no rule on what AI can touch in client work, and often a single person — the owner — as the only reviewer. The risk isn’t the tool; it’s unverified output going out under the business’s name.
Why do small businesses adopt AI without a policy?
Because adoption is easy and policy feels optional. A low monthly subscription removes the cost barrier, the productivity gain is immediate, and writing governance rules takes time the owner doesn’t feel they have. The gap between using AI and using it responsibly is where most small-business exposure lives.
How should a small business owner verify AI-generated work?
Start by naming a reviewer for anything client-facing, even if it’s still you on a scheduled pass rather than in the moment. As an AI ethics speaker and author, I tell owners to write down three rules: what AI may draft, what a human must confirm, and what AI never touches. A one-page standard beats a tool you don’t govern.
Is freed-up time from AI actually a good thing?
It can be, but only if you decide how to spend it. When AI gives an owner back five hours a week and those hours fill with more output and no added review, the time saved becomes risk added. The benefit is real; the discipline to manage it is what makes it pay off.
The tools will keep getting cheaper and faster. What won’t take care of itself is the question of who’s accountable for the work. That’s a human question, not a technical one — and it’s the conversation I have with audiences and leadership teams every week. If your organization is adopting AI faster than it’s deciding who checks the work, let’s build the standard before the consequence finds you. You can reach me at ChuckGallagher.com.
Five Questions for Reflection
- In your business, who reads client-facing work before it goes out — and has that answer changed since you started using AI?
- If AI has given you back time each week, where is that time actually going? Can you name it?
- What would it cost you if one unchecked, AI-generated deliverable reached your most important client with an error in it?
- Do you have a written standard for what AI is allowed to touch in your work, or does that standard live only in your head?
- Which part of your week feels more like efficiency than it should — and what consequence might be hiding inside it?
