New Data: 75% of AI Gains Go to Just 20% of Companies

Two major reports just dropped. Small businesses should pay attention.

PwC and Stanford both released landmark AI research today, April 13, and the numbers tell the same story from different angles. AI adoption is growing faster than the personal computer or the internet. Organizational use has hit 88%. And yet most of the economic value is flowing to a small group of companies at the top.

PwC’s 2026 AI Performance Study found that 74% of all AI-driven economic gains are captured by just 20% of organizations. The Stanford HAI 2026 AI Index, released the same day, confirms that while AI has reached 53% population adoption in three years, the gap between who uses AI and who profits from it keeps widening.

If you run a small business, these reports are not abstract. They describe the competitive landscape you are operating in right now.

What the reports found

PwC: the 75/20 split

PwC surveyed 1,217 senior executives across 25 sectors. The headline finding: AI leaders — the top 20% of companies by AI-driven financial performance — deliver 7.2 times more revenue and efficiency gains than their peers.

The differentiator is not spending. It is strategy. Top performers are 2.6 times more likely to use AI to reinvent their business models, not just trim costs. They are nearly twice as likely to deploy AI in advanced ways — executing multiple tasks within guardrails (1.8x) or operating as autonomous, self-optimizing systems (1.9x).

The bottom 80% of companies are mostly using AI for basic productivity. Spell-checking emails. Summarizing documents. Tasks that save minutes, not thousands.

Stanford: adoption outpaces everything

Stanford’s report tracks the macro picture. Key numbers:

53% population adoption of generative AI in three years, outpacing both the PC and the internet at the same point in their adoption curves
88% of organizations now use AI in at least one business function
$172 billion in annual value to U.S. consumers from generative AI tools, with the median value per user tripling between 2025 and 2026
Employment for software developers aged 22-25 has fallen nearly 20% since 2022, confirming that AI’s workforce impact is already measurable

The U.S. ranks 24th globally in AI adoption at 28.3%, behind Singapore (61%) and the UAE (54%). The tools are available everywhere. The results are not.

Why this matters for small businesses

The divide is not about size

Reading these reports, it is easy to conclude that AI is a big-company game. That would be the wrong takeaway.

PwC’s data shows the divide is between strategic AI users and random AI users — not between large and small companies. The firms capturing 74% of the value are not just throwing money at AI. They are picking specific problems, deploying AI with clear guardrails, and measuring results against revenue, not just productivity.

That is exactly the kind of focused approach small businesses are built for. A plumbing company that uses AI dispatch to route emergency calls and optimize technician schedules is doing what the top 20% does — solving a revenue problem, not just a convenience problem.

The gap we flagged is getting worse

We wrote about this exact dynamic five weeks ago when McKinsey data showed 88% of businesses use AI but only 6% see earnings impacts above 5%. Today’s PwC data confirms and sharpens the picture: the gap is not closing. It is concentrating.

The CFO survey data from last week adds another dimension. Large enterprises are cutting 502,000 roles to fund AI bets that have not yet paid off. Meanwhile, small businesses that already use AI effectively are in a position to pick up the customers and talent those restructurings leave behind.

Appalachia’s position

For businesses in the Appalachian region, the reports carry a specific warning. Rural areas already face adoption barriers — limited broadband, smaller talent pools, tighter budgets. When 75% of AI’s value flows to companies that are already ahead, the risk of falling further behind is real.

But the flip side is equally true. A vacation rental operator in Canaan Valley who uses an AI agent to handle bookings and guest communication is deploying AI the same way PwC’s top performers do: as a growth tool, not just a cost cutter. Scale does not matter when the approach is right.

Our take

What is missing from the headlines

Most coverage of these reports will focus on the competition angle — the U.S. versus China, Anthropic versus OpenAI. That is interesting but irrelevant to a restaurant owner in Charleston or an HVAC contractor in Roanoke.

What matters is what PwC buries in the methodology: the gap is not driven by access to AI or even by budget. It is driven by how companies define the problem they want AI to solve. The top 20% start with a business outcome — more revenue, faster growth, new markets — and work backward to the AI tool. The bottom 80% start with the tool and hope something improves.

The bottom line: AI’s economic gains are concentrating, but the playbook the winners use is available to any business willing to be specific about what it wants AI to do.

Questions that remain

How long does the gap persist? PwC’s data is from mid-2025. As AI tools get cheaper and easier to deploy, the bottom 80% should close ground — but that requires a shift in strategy, not just access.
Will rural and small-market businesses close the gap or widen it? Stanford’s data shows adoption varies wildly by geography. The tools are global. The outcomes are not.

What you should do

Three things to act on this week

Audit your current AI use. Are you using AI for productivity (spell-check, summaries) or for revenue (lead capture, scheduling, customer retention)? If it is all productivity, you are in the 80%.
Pick one revenue problem. Not a general wish for “more efficiency.” A specific problem: missed calls, scheduling gaps, review management, lead qualification. Then find the AI tool that solves it.
Set a 30-day metric. The top 20% measure AI against financial outcomes. If you cannot tie your AI tool to a dollar figure in 30 days, you are not using it strategically.

Watch for

Stanford’s full dataset will be available in the coming weeks, with sector-by-sector breakdowns. Look for the small business and services data.
PwC’s AI fitness index offers a framework for scoring your own AI maturity. It is designed for enterprises, but the principles — particularly around moving from experimentation to production — apply at any scale.

This is the fork in the road

The data is clear. AI is not a future trend. It is a present economic force, growing faster than any technology before it. The question for small businesses is not whether to use AI — 88% already do. The question is whether you are using it the way the top 20% does.

If you are ready to move from general AI use to strategic AI deployment, we can help you figure out where to start.