The global economy is proving more durable than many forecasters expected, but the resilience is increasingly uneven—and dependent on forces that could reverse quickly. The International Monetary Fund’s October outlook projects world growth of 3.1% in 2026, a slight improvement from its July estimate, followed by 3.4% growth in 2027. Yet the expansion remains below 2025’s 3.5% pace, while global inflation is expected to rise to 4.7% this year before easing to 4.1% next year. The IMF’s forecast[1]
The headline numbers conceal a more consequential story: artificial-intelligence investment is helping offset the economic damage from war and higher energy costs, but the benefits are concentrated in countries and companies already positioned inside the technology supply chain. Energy importers, lower-income economies and governments carrying heavy debt loads face a less forgiving environment.
Two shocks moving in opposite directions
The IMF describes the current outlook as the interaction of two contrasting shocks. Conflict in the Middle East has lifted energy and other commodity prices, functioning as a negative supply shock that raises costs while suppressing activity. At the same time, the AI investment cycle has created a positive demand shock, supporting trade, capital spending and selected technology industries. IMF briefing[2]
This helps explain why global growth has not fallen more sharply. Semiconductor producers, data-centre builders, cloud companies and economies exporting technology-related goods can benefit from the surge in spending. Countries with favorable commodity terms of trade may also absorb higher prices better than import-dependent states.
But the same pattern creates a widening divide. Higher fuel and food bills hit households and businesses directly, leaving poorer economies with less room to subsidize consumption or protect investment. For governments, higher interest rates and rising bond yields increase the cost of refinancing debt just as budgets are being pressured by energy support, defense spending and weak domestic demand.
Why the AI boom is both support and risk
The optimistic case is that today’s investment will eventually produce broad productivity gains. If AI allows firms to automate routine work, improve logistics and develop products faster, the initial spending could translate into stronger potential growth rather than merely higher asset prices. The IMF identifies faster-than-expected productivity gains as an upside scenario. World Economic Outlook[1]
The less reassuring possibility is that expectations have outrun results. A large share of the boom depends on continued capital expenditure by a relatively small group of technology companies and their investors. If projected revenues fail to justify current valuations, investment could slow, markets could correct and demand would weaken beyond the technology sector. The IMF warns that a reversal in AI enthusiasm could leave global output below its baseline, while exposure would be particularly significant in the United States. Axios analysis[3]
That does not mean AI investment is necessarily a bubble. Supporters point to strong demand for computing capacity and the possibility that productivity improvements will spread gradually, as earlier general-purpose technologies did. Skeptics counter that the gains remain difficult to measure and may arrive later than markets assume. Both views imply volatility: the economic benefit depends not only on innovation, but on whether investment remains financially sustainable.
What policymakers can—and cannot—control
The immediate policy challenge is balancing inflation against weak demand. Central banks cannot produce more oil or end a conflict, but they can prevent temporary energy shocks from becoming embedded in wages and prices. Moving too aggressively to support growth risks prolonging inflation; keeping policy tight for too long risks worsening debt stress and unemployment.
Governments face an equally difficult choice. Broad energy subsidies can cushion households but strain public finances and reduce incentives to conserve energy. Targeted assistance is cheaper, yet harder to administer and politically less popular. Lower-income countries may need external financing and debt restructuring to preserve spending on health, education and infrastructure rather than diverting it toward emergency imports and interest payments.
Trade policy will also shape the outcome. Renewed tariff disputes could interrupt the very supply chains supporting the technology boom, while restrictions on advanced chips and related equipment may deepen the divide between countries inside and outside the AI ecosystem. Strategic competition may encourage domestic investment, but it can also make production more expensive and reduce efficiency.
The next test
The IMF’s central forecast assumes that geopolitical pressure does not intensify dramatically and that AI investment continues to support activity. That is a narrow path. An escalation that keeps energy prices elevated would compound inflation and weaken importers; a sharp AI market correction would remove an important source of demand; and higher sovereign yields could expose fiscal vulnerabilities that currently remain manageable.
The most plausible near-term outcome is therefore not a uniform global downturn, but a split economy: resilient technology and commodity exporters alongside countries facing higher bills, weaker currencies and constrained public budgets. The growth rate may look stable from a distance. For policymakers and households, the more important question is who is receiving the growth—and who is paying for its risks.