NextFin News - Artificial intelligence is starting to do for the rest of the world what it has done for the United States: lift economic growth. The International Monetary Fund says the AI investment boom that powered the American economy is now spreading beyond its borders, pulling up countries integrated into the global technology value chain and adding enough momentum to keep global expansion resilient even as war in the Middle East and trade fragmentation weigh on the outlook. The catch, in the Fund's own telling, is that most of the lift so far is a demand-side investment wave — not yet the productivity payoff that would make the gains permanent. And the latest numbers show just how conditional that promise remains.
The Situation: A Broadening Boom, A More Cautious Forecast
The IMF's July 2026 World Economic Outlook Update, titled "Global Economy in Crosscurrents of War and Technology," projects global growth at 3.0 percent in 2026 and 3.4 percent in 2027, down from the 3.5 percent average recorded in 2024-25. The aggregate picture is "broadly unchanged" from the April forecast, but the composition has shifted in a way that matters: the Fund explicitly credits "AI-driven demand" for lifting countries tied into the global technology value chain, while commodity importers that cannot plug into AI-driven activity face downgrades. Growth in advanced economies is projected at 1.7 percent in 2026 and 1.8 percent in 2027, with net energy importers facing a more pronounced drag from higher energy prices "unless they are lifted by technology-related activity."
This is the second act of a story that began earlier in the year — and the two vintages of the forecast tell different parts of it. In January, the IMF had raised its 2026 global growth forecast to 3.3 percent, up 0.2 percentage points from October 2024, and pinned a large part of the American outperformance on massive investment in AI infrastructure: data centers, high-end chips, and the power grid to run them. The United States' 2026 growth forecast was lifted to 2.4 percent, a 0.3 percentage point upgrade, before the Fund trimmed its 2027 view to 2.0 percent on the logic that the investment surge would not repeat at the same pace. At that time, the Fund estimated that if the AI investment surge translated into rapid adoption and real productivity gains, global growth "may be lifted by as much as 0.3 percentage points in 2026 and between 0.1 and 0.8 percentage points per year in the medium term, depending on the speed of adoption and improvements in AI readiness globally."
Six months later, the July update is more cautious — global growth now at 3.0 percent for 2026, the United States at 2.3 percent — because war-related energy shocks and trade fragmentation have intervened. But the mechanism the Fund identified in January has not gone away; it has broadened. Chief economist Pierre-Olivier Gourinchas has described the AI boom as "sucking in capital from around the world," with foreign investors eager for a share of the returns. The wealth effect is no longer confined to US stock markets: the Fund has pointed to countries such as South Korea, where AI-linked valuations and export demand are feeding domestic activity. In a video interview, Gourinchas put the transmission plainly: the boom is "boosting economies in Asia that are very plugged into the AI tech" supply chain.
The clearest evidence sits in the most recent outcomes. South Korea, despite heavy reliance on imported energy from the Middle East, recorded a 7.5 percent growth rate in 2025 — more than four times the 1.8 percent projected in the April 2026 forecast — powered primarily by a semiconductor and AI-hardware export boom. China expanded at 8.1 percent on the Fund's seasonally adjusted estimates, driven by front-loaded public infrastructure investment and a surge in high-tech activity. In Europe, technology investment helped offset tepid consumption amid weak household confidence. Earlier in the year, Spain earned a 0.3 percentage point upgrade to a 2.3 percent 2026 forecast on the back of technology investment, while Britain was left unchanged at 1.3 percent.
The stakes are quantified, and the range is the whole story. The difference between a modest tailwind and a transformation is not the size of the investment — it is how fast the rest of the world can actually use what is being built. The IMF's 0.1-to-0.8 percentage point range turns on exactly that: adoption speed and AI readiness.
"We find that global growth remains quite resilient," Gourinchas told reporters in January. "So, in a sense, the global economy is shaking off the trade and tariff disruptions of 2025 and is coming out ahead of what we were expecting before it all started."
But resilience is not the same as durability. And that distinction is where the analysis has to go.
The Mechanism: Capital First, Productivity Later
The first thing to get right is the channel. The AI boom is reaching the rest of the world through two distinct mechanisms, and they operate on different clocks.
The first is a capital-flow and demand channel. Hyperscalers and chip designers, concentrated in the United States, are spending at a pace that pulls in suppliers from Taiwan, South Korea, Japan, and increasingly Europe. The Bank for International Settlements estimates the five largest hyperscalers alone are set to spend more than $1 trillion on AI-related capital expenditure from 2025 through 2026 — commitments that are outpacing earnings and free cash flow, forcing some firms to issue debt to finance the buildout. Every dollar of that spending does not stay in California. It flows to semiconductor fabrication, data storage, servers, and digital infrastructure wherever those supply chains sit. That is why Asia, already wired into the technology value chain, is seeing the first and largest spillover — and why world trade volume growth is projected to slow sharply from 5.0 percent in 2025 to 3.5 percent in 2026 before recovering to 4.3 percent in 2027, with the brisk growth of technology-related trade flows partly offsetting tariff drag.
The second channel is productivity diffusion, and it runs slower. For AI to lift trend growth outside the United States, firms in Europe, Asia, and emerging markets must actually deploy the technology, reorganize work around it, and capture efficiency gains. The IMF's own scenario work finds that advanced economies — particularly the United States, home to the major AI service providers — capture a disproportionate share of AI-related gains, while emerging markets and low-income countries gain less because of weaker infrastructure and limited capacity to deploy AI technologies. The Fund's scenario planning shows persistent output gains only under a "runaway diffusion" path in which confidence in and social acceptance of AI technologies support broad deployment.
So the near-term lift is cyclical demand; the long-term prize is structural productivity. Confusing the two is the most common error in reading this moment.
Cyclical Versus Structural: Two Forces, Not One
This is a cyclical investment boom riding on top of a structural technological shift, and they point in different directions.
The cyclical leg is the capex super-cycle itself. It has the classic markers: a concentrated group of buyers committing to multi-year spending plans, suppliers ramping capacity, and financing conditions accommodative enough to let spending run ahead of current earnings. Cyclical booms of this kind revert when the expected return on the installed base fails to materialize. History is littered with them — the fiber-optic buildout of the late 1990s, the commodity super-cycle of the 2000s, the shale boom of the 2010s. Each lifted growth and trade while it lasted; each left overcapacity when demand disappointed. The evidence floor for calling this leg cyclical is met: there is a short-term driver in hyperscaler capex commitments, a clear supply-chain transmission channel, and a demonstrated pattern of mean reversion in prior technology capex cycles.
The structural leg is different. If AI proves to be a general-purpose technology that diffuses broadly, it changes the production function itself — the same way electrification did in the early 20th century. That would not revert on its own. The evidence for this leg is thinner but real: AI-related investment already accounts for a large share of GDP growth in the advanced economies, fueling demand for servers, data centers, software, and power infrastructure, and the Fund's scenario planning shows persistent output gains under a runaway-diffusion path. A structural claim requires evidence of a permanent regime change; here it rests on whether deployment becomes broad-based rather than confined to a handful of tech giants.
The correct read, then, is to separate the legs. In the short run, the world is riding a US-led investment wave that is spilling into Asia's supply chains and, more modestly, into European capital spending. In the long run, the question is whether that investment converts into economy-wide productivity — and on that question the IMF is explicitly non-committal, which is the honest position.
The Second-Order Effect the Market Is Not Pricing
The consensus view is straightforward: more AI investment means more growth, especially outside the United States. The second-order question is harder, and it cuts the other way.
A US-centered AI boom that "sucks in capital from around the world" strengthens the dollar and lifts US asset prices. That is good for American wealth and consumption. For the rest of the world, the immediate effect is a mix: Asian exporters gain order books, but capital outflows toward US AI champions can tighten financial conditions elsewhere and push up the cost of financing the very AI infrastructure those countries want to build. The wealth effect Gourinchas flagged — rising valuations in the US and South Korea feeding consumption — is a double-edged transmission. It supports demand today; it also concentrates the gains in the economies that already own the AI assets.
There is a further step. If the productivity payoff arrives slowly, the investment wave will have to be financed by debt and equity issuance for longer than earnings can justify. The Bank for International Settlements has already warned that hyperscaler commitments are outpacing free cash flow, with some firms turning to debt markets. Should investor patience thin, the correction would not be confined to technology shares. A retrenchment in AI capex would hit the Asian supply chain first and hardest — the same economies the IMF now credits with the strongest surprise growth — and through them, commodity exporters and European equipment makers. The boom's greatest beneficiaries are also its most exposed.
The Counter-Thesis: What If the Payoff Never Arrives?
The strongest argument against the upbeat reading comes from the Fund's own risk accounting and from the Bank for International Settlements. The concern is not that AI is unimportant; it is that the market is pricing the diffusion scenario while the economy is still living through the investment scenario.
The July 2026 update frames the technology shock as "more distinctly two-sided." On the one hand, if the recent surge in AI-related investment were to translate more rapidly into broad-based deployment and efficiency gains, medium-term growth could strengthen. On the other, a reassessment of AI profitability and productivity expectations could bring "tighter global financial conditions, balance sheet pressures, and weaker activity extending beyond the technology sector." The Bank for International Settlements puts the same risk in harder numbers: the five largest hyperscalers' trillion-dollar capex commitments are running ahead of earnings and free cash flow, and the investment race is being financed partly by debt. If AI-driven profitability and productivity gains are revised downward, it warns, "investment in technology-intensive sectors could retrench," with weaker activity spreading beyond technology.
This counter-thesis attacks the core of the bullish case at its foundation — the assumption that investment will convert into productivity. It is backed by the two institutions whose job it is to stress-test the global outlook, and it deserves more than a passing mention.
The falsifying signal is specific. Watch the ratio of AI-related revenue growth to AI capital expenditure among the largest hyperscalers over the next four quarters. If capex keeps climbing while revenue per dollar of capex fails to rise — that is, if the marginal return on AI investment does not improve — the structural-productivity thesis is wrong, and the cycle will revert through a capex retrenchment rather than a productivity step-up. A secondary signal: if global growth undershoots the Fund's own estimated AI uplift of up to 0.3 percentage points in 2026 even as investment stays high, the diffusion is not happening.
Outlook: Who Benefits, Who Is Exposed
The bottom line is that AI is becoming a genuine multi-region growth engine, but the engine is running on investment demand, not yet on productivity. That distinction determines who benefits, who is exposed, and for how long.
In the short term — the next few quarters — the beneficiaries are clear: economies embedded in the AI hardware supply chain. South Korea's semiconductor exporters, Taiwan's foundries, Japan's equipment makers, and the US hyperscalers and chip designers themselves are riding the steepest part of the curve. Commodity exporters tied to energy demand for data centers also gain. The exposed are the laggards: commodity importers without AI readiness, and economies whose growth models depend on cheap capital that a US-centered boom may draw away.
Over the medium term, the split widens. If adoption accelerates and AI readiness improves, the 0.1-to-0.8 percentage point annual uplift becomes real, and the gains broaden to services and non-tech sectors in Europe and parts of Asia. If adoption stalls, the investment wave peaks and the countries most dependent on AI capex orders face the sharpest slowdown. The IMF's base case — 3.0 percent global growth in 2026, recovering to 3.4 percent in 2027 — effectively prices the middle of that range, with war-related energy shocks and trade fragmentation as the offsetting drags.
Three signals will decide which path the world takes. First, the hyperscaler revenue-to-capex ratio named above — the single best gauge of whether the investment is earning its keep. Second, the pace of AI adoption outside the technology sector, which the IMF ties directly to the size of the growth uplift. Third, financial conditions: if debt-financed AI spending forces a repricing in bond or equity markets, the demand-side lift reverses quickly.
The IMF's message is not a celebration of a new era. It is a conditional one: AI can fuel global growth as investment spreads beyond the United States, but only if the rest of the world builds the capacity to use it. The investment is already global. The productivity is not.
The market is being asked to believe in the second half of that sentence while the evidence only supports the first.
Explore more exclusive insights at nextfin.ai.
