The K-Shaped AI Economy: Why Southeast Asia's Future Rides on Its Small Businesses

When people picture the future of Southeast Asia's economy, they tend to imagine gleaming corporate towers and billion-dollar tech champions. But that's not where most of the growth actually happens. Our region runs on small businesses — the neighbourhood eatery, the online seller shipping from a spare room, the family workshop that's been going for three generations.

That's why the biggest economic story of the next few years isn't which giant builds the best AI. It's whether our smallest businesses can use AI at all. And right now, the answer is splitting in two — a K-shape, where large firms race ahead and small ones drift behind.

Small businesses are the engine, not the sideshow

Small and medium enterprises are the backbone of Asia's economy. They create most of the jobs, drive a huge share of output, and hold communities together. When an SME does well, a whole street does well — suppliers, staff, families.

In payments, we see this up close every day. The merchants tapping QR codes, the sellers reconciling e-wallet payouts, the shop owners checking yesterday's takings on their phone — that's the real economy, transaction by transaction. If we want the region to thrive, these are the businesses that have to thrive.

The new gap isn't digital. It's AI.

Here's the good news: on basic digital skills, small businesses have largely caught up. A decade ago, many were cash-only. Today they accept QR payments, sell online, run a delivery account, and message customers on chat apps. The digital divide we worried about has narrowed a lot.

But a new gap is opening, and it's different. It's not about coding or building complex models. It's about AI literacy — simply knowing how to put AI tools to work: drafting a supplier email, forecasting next month's stock, spotting a suspicious transaction, answering customers around the clock.

This is a softer skill than "learn to program," but it's just as decisive. And it's exactly where small businesses are falling behind.

The divide is widening — and that's the danger

The numbers are stark. Fewer than 2% of SMEs in Asia have AI-related skills today — four to five times lower than large enterprises. And the gap isn't closing. It's widening.

That's the shape of a K. Large firms have the budgets, the specialists, and the time to fold AI into everything they do. They get faster and cheaper every quarter. Small firms, busy just keeping the lights on, don't get that head start — so the distance between the two keeps growing. Left alone, a K-shape doesn't self-correct. The top arm and the bottom arm only pull further apart.

A skills gap becomes a resilience gap

Here's why this should worry all of us, not just the businesses concerned. As AI gets baked into everyday work — pricing, marketing, service, fraud checks — the AI gap stops being a "nice to have" skills issue and becomes an economic resilience issue.

If the businesses that employ most of the workforce can't keep pace, the whole economy gets more fragile. Productivity concentrates in a few large players. Local jobs get squeezed. Whole sectors become less able to absorb the next shock — a supply disruption, a downturn, a new competitor. A divide that starts on a laptop ends up on the region's balance sheet.

The flip side: AI is the great leveller — if we let it

Now the hopeful part. The same technology that threatens to widen the gap is also the best chance small businesses have ever had to level up.

AI lowers the old barriers to entry. Insights that once needed a data team, marketing that once needed an agency, customer service that once needed a call centre — a small business can now get a workable version of all three from tools that cost very little. For the first time, the corner shop can reach for capabilities that used to belong only to the big chain. The opportunity is real. The question is whether small businesses get the tools — and the confidence to use them — before the divide hardens.

What the payments and fintech ecosystem should do

This is where our industry has an unusual advantage, and an unusual responsibility.

Banks, fintech platforms, and large merchants already reach small businesses at a scale no training programme ever will. We process their payments, hold their accounts, lend them working capital, and sit on the transaction data that could power genuinely useful AI for them. Just as important, small businesses trust their bank or payment provider far more than a random app they've never heard of.

That gives the ecosystem a clear job: make AI adoption invisible. Don't ask a busy shop owner to become an "AI user." Bake the intelligence into the tools they already use — cash-flow forecasts in the app they check each morning, fraud alerts on the terminal, demand predictions in the dashboard, a simple assistant that drafts their invoices. Meet them inside the workflow, not in a classroom.

And no one can do it alone. Closing this gap takes deliberate, coordinated investment — in infrastructure, in accessible tools, in partnerships, and in plain-language skills. Governments set the direction and fund the reskilling. Industry bodies spread standards and good practice. Schools and training providers build AI literacy from the ground up. And businesses — especially those of us in the payments ecosystem — put the tools directly into small hands.

Two shapes, one choice

Southeast Asia is standing in front of two very different shapes. One is a K — a region that splits into a fast lane for the few and a slow lane for the many. The other is a rising tide, where AI lifts the millions of small businesses that already power our economy.

The technology won't decide which shape we get. We will. And in payments, we sit closer to these businesses than almost anyone — which means we don't get to sit this one out.