The global economy is displaying an unusual combination of strength and fragility. The International Monetary Fund expects world output to grow 3.1% in 2026 and 3.4% in 2027, a modest improvement on its previous forecast for this year but a slowdown from 3.5% growth in 2025. At the same time, inflation is projected to rise to 4.7% in 2026 before easing to 4.1% next year. The figures suggest that the world economy is absorbing major shocks without falling into recession—but also that its apparent resilience depends increasingly on a narrow set of forces.
The IMF’s October World Economic Outlook identifies two opposing pressures. Higher energy and food prices linked to war are acting as a negative supply shock, raising costs and restraining production. Artificial-intelligence investment, by contrast, is generating a positive demand shock through technology spending, trade and capital expenditure. The result is not a uniformly healthy expansion, but an uneven one in which some economies benefit from the technology cycle while others absorb its inflationary and geopolitical costs.
Why the headline numbers conceal a divide
Aggregate growth can obscure sharply different national experiences. Countries with companies, supply chains or financial markets tied to AI investment are receiving a boost from data centers, semiconductors, software and related infrastructure. Economies benefiting from favorable terms of trade may also withstand higher energy prices better than import-dependent states.
The IMF says the outlook is becoming increasingly uneven. Its analysis indicates that many energy importers and countries exposed to conflict face downgrades, while economies more closely linked to the AI boom have been upgraded. That distribution matters politically: a global expansion can continue while households in vulnerable countries face higher transport, heating and food bills.
Energy markets are central to the tension. The IMF describes the war-related increase in energy and commodity prices as a supply shock that simultaneously raises inflation and weakens activity. Recent reporting has also pointed to disruptions involving Middle Eastern fuel supplies and the Strait of Hormuz, a major energy transit route. In the eurozone, energy-price increases contributed to inflation rising to 3.8%, according to Le Monde.
AI: productivity breakthrough or concentrated risk?
The optimistic case is straightforward. If AI improves productivity faster than expected, businesses could produce more with fewer resources, investment could remain strong and higher efficiency could eventually support wages and living standards. The IMF explicitly identifies earlier-than-expected productivity gains as an upside scenario.
But the same report highlights the vulnerability of relying on that prospect. Technology-related investment is becoming concentrated among a small number of firms, sectors and economies. If investors reassess optimistic expectations, a sharp correction in AI-linked assets could weaken business spending and consumer confidence. The IMF also warns that the financial system may be exposed to stretched valuations and the increasing importance of a narrow group of companies.
“Risks surrounding AI are two-sided,” IMF Research Department official Deniz Igan said in the fund’s annual-meetings briefing. The technology could generate faster productivity gains, but expectations could also prove too optimistic.
This is the key uncertainty behind the forecast. AI investment is currently helping offset the drag from energy shocks, but an investment boom is not the same as broad-based productivity growth. The benefits depend on whether firms deploy the technology widely, whether workers can adapt and whether gains reach economies outside the main technology centers.
What policymakers face next
Central banks must navigate an uncomfortable combination of weaker underlying activity and renewed inflation pressure. Cutting interest rates too quickly could allow energy-driven inflation to become entrenched; keeping rates high for too long could damage investment and employment. Governments face a parallel dilemma: they need to protect households from energy and food shocks without worsening already elevated public debt.
The IMF’s message is therefore less a declaration of safety than a warning about diminishing room for error. Growth remains positive, but the expansion is increasingly dependent on AI-related investment while exposed to war, energy disruption and financial-market repricing. A prolonged oil shock could be especially damaging for highly indebted economies, according to the fund’s latest assessment.
The next phase will test whether resilience is genuine or merely statistical. If energy markets stabilize and AI delivers measurable productivity gains, global growth could accelerate in 2027. If conflict-driven costs persist or technology valuations fall sharply, the same concentration that is supporting activity could amplify the downturn. The decisive question is whether governments can broaden the benefits of the expansion before its risks become more widely shared.