Publication date: 13 October 2026
The International Monetary Fund has raised its forecast for global growth in 2026 to 3.1%, from 3.0% in July, and expects 3.4% growth in 2027. That modest upgrade looks reassuring. The distribution beneath it is not.
The IMF’s October World Economic Outlook describes an economy supported by heavy investment in artificial intelligence while conflict, expensive energy and high public debt erode resilience elsewhere. Global inflation is forecast at 4.7% this year and 4.1% in 2027, keeping monetary and fiscal choices difficult even as headline output holds up.
Growth is becoming more concentrated
AI infrastructure is doing more than lifting technology shares. Spending on data centres, computing equipment and related supply chains is feeding directly into investment and trade. The IMF has upgraded several advanced economies and hardware exporters that sit inside this cycle.
But the same mechanism creates a widening gap. Reuters reported that 32 countries now account for 95% of global growth, compared with 54 countries in the IMF’s April forecasts. Nearly half of the economies covered have been downgraded since April. Energy importers outside the AI value chain face the hardest combination: higher input costs without an offsetting investment boom.
The headline forecast therefore risks disguising concentration. A global average can improve even while a larger number of countries deteriorate, provided a small group of large economies performs strongly enough.
Three risks behind the upgrade
First, the outlook assumes the energy disruption will eventually ease. The IMF’s projections are built around the Strait of Hormuz reopening in March 2027 and shipping conditions broadly normalising by October 2027. A more severe scenario, with oil at $130 a barrel through the first quarter of 2027, would reduce growth and lift inflation further.
Second, higher sovereign-bond yields increase the cost of refinancing public debt. Countries with limited fiscal space may have to choose between supporting households, servicing debt and funding long-term investment. That makes temporary, targeted help more sustainable than broad subsidies, but politically harder to design.
Third, the AI boom is itself a two-sided risk. Faster adoption could raise global output, but disappointing productivity could reverse investment and asset prices. The IMF estimates that a sharp reassessment could weaken US output directly and spill into Asian technology exporters, with tighter financial conditions amplifying the damage.
A resilience test for leaders
The practical message is not that AI investment is artificial or unimportant. It is that resilience built on one capital-intensive cycle is narrower than broad-based growth. Governments need fiscal buffers, energy security and access to the skills and infrastructure required to use new technology productively.
For business leaders, the same lesson applies. Demand tied to the AI build-out may remain strong, but exposure to energy costs, borrowing rates and concentrated customers deserves as much attention as the growth headline. The world economy is still expanding. It is doing so on a smaller and less forgiving base.