AI as an Operating Tool, Not a Magic Trick

Most of the conversation about AI in business swings between two extremes. On one side, it will replace entire departments next quarter. On the other, it's a toy that writes mediocre emails. Neither view is useful to someone who has to run a company on Monday morning.

I think about AI the way an operator thinks about any tool: what job does it do, how reliably, at what cost, and what has to be true in the business for it to help rather than add noise?

AI amplifies the system you already have

If your business has clear priorities, defined processes, and clean information, AI can make that system faster and more visible. If your business runs on tribal knowledge, scattered spreadsheets, and unclear ownership, AI will mostly help you produce confusion more efficiently.

That's not a reason to wait. It's a reason to treat AI adoption as an operations project rather than a technology purchase. The work of getting ready — documenting how things actually get done, deciding who owns what, getting key numbers in one place — pays off whether or not you ever deploy a model.

Where AI tends to earn its keep

• Summarizing and surfacing. Turning meeting notes, job updates, and reports into a clear picture of what needs attention.
• First drafts. Proposals, procedures, job descriptions, and plans that a knowledgeable person then refines.
• Consistency. Applying the same checklist or standard across many situations without fatigue.
• Connecting dots. Helping leaders see how a decision in one area might affect priorities, budget, or people elsewhere.

Where judgment still belongs to people

AI is good at patterns and poor at accountability. It can suggest; it cannot own. Decisions about people, customers, safety, and capital need a human who understands the context and will stand behind the call. A useful rule: let AI prepare the decision, and let a person make it.

Let AI prepare the decision. Let a person make it.

There are also real risks to manage — data privacy, inaccurate outputs stated with confidence, and the temptation to automate a broken process instead of fixing it. Good operators treat these the way they treat any operational risk: name it, put controls around it, and review it regularly.

Why I'm building ABOS

This thinking is the reason I started working on ABOS, an AI business operating system that is currently in development. The idea is simple: help leaders connect strategy, priorities, and daily execution in one place, with AI assisting the cadence rather than replacing the people who run it. It's early, and I'm building it carefully, but the problem it addresses is one I've lived with for my entire career.

A practical starting point

If you're an operator wondering where to begin, pick one recurring process that eats leadership time — a weekly report, a status meeting, a review — and ask how AI could make it faster and clearer without removing human ownership. Run it for a month. Measure the time saved and the quality of the decisions. Then decide what's next.

AI won't run your business. But used well, it can help you run it with more clarity than ever before.