Vanguard forecasts 3% U.S. GDP growth in 2027, a call published by The Japan Times on August 26, 2026. Joseph H. Davis argues the gains hinge on artificial intelligence advancing from simple task replacement to tools that lift workers and spur new products. If that happens, the winners in an AI-supercharged economy won’t just be the model makers. They’ll be the users who reshape how work gets done.
What Vanguard’s call signals for an AI-supercharged economy
According to The Japan Times commentary by Davis, Vanguard’s higher growth outlook implies stronger support for risk assets. The view rests on a shift in AI’s role: from automating narrow tasks to augmenting human decision-making, then enabling entirely new services and industries. That arc mirrors past waves of general-purpose tech, where the long run payoff arrived once businesses redesigned workflows, not when they first bought the tool.
The near-term takeaway is practical. If companies push past pilot projects and wire AI into daily operations, productivity improves without a one-to-one staff cut. In the medium term, firms start to launch offerings that weren’t viable before. That’s where profits widen and customer surplus grows. Vanguard’s thesis, as presented in The Japan Times, is that the biggest share of value accrues during those later phases.
Investor frenzy for builders vs. real‑economy gains
Capital today is pouring into model developers and core infrastructure. On August 26, 2026, the South China Morning Post reported that Hangzhou-based DeepSeek is nearing a funding round valuing the startup at about US$74 billion, ahead of a planned Star Market listing as early as 2027. Investors span Chinese venture funds and industrial players, underscoring how much attention the AI “builders” command.
That enthusiasm is rational for now: models and compute remain bottlenecks. But history suggests the profit pool broadens as adoption deepens. In the personal computer and internet eras, outsized gains eventually accrued to firms that reorganized their businesses around the tech — retailers that mastered e‑commerce logistics, manufacturers that embedded software into products, and service providers that changed how they serve customers. AI looks set to rhyme with that pattern. The builders spark the boom; the users decide how large it becomes.
Where value accrues in an AI‑driven economy
Early studies offer a clue. A 2023 working paper by Erik Brynjolfsson and co‑authors found that access to a generative AI assistant boosted the productivity of customer support agents by 14% on average, with the biggest gains for novices (NBER, May 2023). That’s augmentation in action: the same workforce handles more, and quality rises. It aligns with Davis’s point in The Japan Times that the next gear comes when AI makes workers better, not just cheaper.
Who benefits as that pattern spreads? Three groups stand out:
- Operational laggards with clean data. Firms that were late to digitize often carry higher costs. They can realize step‑changes quickly by pairing AI with disciplined data work. The margin impact can be meaningful even before new products arrive.
- Service industries with repeatable knowledge work. Contact centers, insurance operations, accounting, and parts of healthcare can combine retrieval, summarization, and guided workflows to cut handling times and errors. Gains compound as more processes get re‑written.
- Companies ready to ship new bundles. When AI enables features customers actually pay for — faster resolution, self‑service planning, predictive maintenance baked into a contract — the value shows up in pricing power, not just cost savings.
There’s also a labor angle. Augmentation tends to compress performance gaps. New hires climb the learning curve faster, which can shift training budgets and wage structures. That doesn’t mean broad job losses. It means different mixes of skills and higher output per hour. The OECD’s AI Policy Observatory has tracked how diffusion — access, skills, and organizational change — drives whether these micro gains add up to macro growth.
Reading the gap between markets and the street
So why does the market still favor builders? Scarcity and narrative. Core model development, GPU supply, and top-tier AI talent are constrained, which supports high valuations. The user side is messier. It demands change management, data plumbing, and patience — the unglamorous work that rarely excites investors until the numbers move.
This is the crux of the Japan Times argument: if growth does step up, the lift won’t come only from model breakthroughs. It will come from widespread adoption that reconfigures thousands of processes. That’s how an AI-supercharged economy translates into higher aggregate output — many modest improvements compounding across sectors, with a few breakout products layered on top.
Expect uneven timing. Regulated fields will move in fits and starts as guardrails mature. Consumer‑facing wins may arrive faster, since customer experience improvements are easy to ship and measure. Industrial settings will follow as reliability rises and vendors embed AI into equipment and contracts.
What to watch next: adoption, margins, and wage effects
For executives and investors trying to separate signal from noise, a short dashboard helps:
- Adoption depth, not slideware. Track the share of workflows with AI in production, measured by users per day and task completion — not just licenses sold.
- Unit economics. Watch gross margin and service-level metrics where AI pilots scale. If costs fall while satisfaction improves, value is accruing to users.
- Wage and training patterns. Faster ramp times and new pay bands for AI‑assisted roles indicate augmentation, the condition Vanguard says precedes a growth step‑up.
- New revenue lines. Count features customers pay for, not demos. Real pricing power signals the move from automation to new offerings.
Policy also matters. Clear rules on data use, disclosure, and safety lower adoption friction. That’s a precondition for the broad diffusion The Japan Times piece identifies as necessary for a higher growth path. Absent that, AI risks staying bottled up in a few tech balance sheets.
The market may keep rewarding model makers while scarcity lasts. But if Davis’s thesis proves right, the durable gains will accumulate where AI meets process and product. In that sense, the next phase of an AI-supercharged economy won’t be won by whoever ships the biggest model. It will be won by the companies — large and small — that turn those models into better service, fatter margins, and loyal customers. For more on this, see bloomberg.com and nytimes.com.
Related reading: NVIDIA • Meta AI • AI & Big Tech