HERNDON, Va. — For most small-business owners, the conversation about artificial intelligence has sounded the same for a few years now: Here are some tools. Here’s what they can do. Here’s why you should care.

That conversation has changed.

“Back when ChatGPT was new, it was all about tools,” says Charlie Morris, chapter co-chair and marketing and technology committee chair for SCORE Greater Philadelphia. “Now it’s really focused on the outcomes I want to get. What business problem am I trying to solve?”

Morris and fellow SCORE mentor Matthew Krieger recently presented a webinar called “AI in 2026: What’s Changed and What It Means for Small Business” as part of SCORE’s National Small Business Week Learning Lab. Their message: AI has matured past the point of novelty. The question today is not whether it works, but where to put it to use.

Morris, a retired software developer with a master’s degree in computer science concentrating in artificial intelligence, says that, not long ago, dozens of tools were competing for attention, and the pressure was to pick the right one. Now, he says, the tools have largely become commodities.

“There’s an explosion in technology options,” Morris says. “There are dozens and dozens of tools. And it’s really, frankly, what business problem am I trying to solve?”

The biggest development he points to is the move toward AI agents, which can execute multi-step tasks autonomously rather than simply responding to a question. But before getting there, Morris wanted to demonstrate something more immediate: what AI can do with the data a business owner already has sitting around.

Most business management tools are built for clean, structured information. A spreadsheet needs rows and columns. A database needs defined fields. But the information small-business owners actually work with rarely looks like that.

“Spreadsheets need clean, structured rows,” Morris says, “but what owners actually have is a pile of unstructured stuff — emails, point-of-sale exports, receipts, handwritten notes.”

That gap between what the tools expect and what owners actually have has always been a problem. Morris argues AI is the first tool built for the data owners actually possess, not the data they wish they had.

To show what he means, Morris ran a live demonstration using simulated data from a fictional jewelry store — four files in all: a point-of-sale export, customer emails, the owner’s handwritten notes and Google reviews. He dropped them into Claude, an AI tool from Anthropic, and asked a single plain-English question: “What’s going on in my business right now?”

The result, he says, was the kind of cross-referenced analysis that would otherwise take hours. The AI flagged that gold prices had risen 18% but the store hadn’t adjusted its pricing. It noted that the goldsmith was booked out six weeks with walk-in customers being turned away. It spotted that Mother’s Day was nine days out and nothing had been done to promote it.

“It said, ‘Here’s what’s actually going on — your business is healthy, but you’re bleeding on three fronts,’” Morris says.

The demonstration didn’t stop at the overview. Morris showed how a business owner could then drill into any one of those problems. When he asked what the options were for repricing gold inventory, the AI returned a breakdown of four approaches, with tradeoffs for each.

When he asked for a Mother’s Day marketing plan that could realistically be executed in eight days, it produced a timeline covering an Instagram post, an email to past customers and an in-store display table.

The takeaway Morris drew from the demo was deliberate. The fictional store owner hadn’t hired a consultant, built a dashboard, or taken a data analytics course. She took four files she already had, put them in one place, and asked a plain question.

“AI didn’t run her business,” Morris says. “It helped her see more clearly than she could on her own.”

For dry cleaners, the parallel is direct. Customer emails, point-of-sale reports, online reviews, staff notes — the data is already there. The question is whether it’s being put to use.

“What four files do you have sitting around that are trying to tell you something?” Morris asks.

Come back Thursday for Part 2, where we’ll look at using AI as a thought partner for difficult business decisions, and how agents can handle multi-step tasks on your behalf.