The global economy is being pulled in opposite directions. Artificial-intelligence investment is driving trade, capital spending and financial-market optimism, while conflict in the Middle East is pushing up energy prices and weakening household and business confidence. The result is not a conventional expansion or recession story, but an increasingly fragile balance between a technology-led demand surge and a war-driven supply shock.
The International Monetary Fund projects global growth of 3.1% in 2026, down from 3.4% last year, followed by a modest acceleration to 3.4% in 2027. It expects headline inflation to rise to 4.7% this year before easing to 4.1% next year. Those figures suggest that the world economy is still expanding, but with less room to absorb another shock. (IMF World Economic Outlook)
Two shocks, moving in opposite directions
Energy prices are transmitting the conflict’s effects far beyond the countries directly involved. Higher oil and other commodity prices raise transport, production and food costs, squeezing consumers while forcing central banks to weigh inflation against weaker growth. Reuters reported that Brent crude rose above $117 a barrel amid renewed uncertainty over diplomacy between the United States and Iran. (Reuters)
That pressure is particularly severe for energy-importing and lower-income economies. Governments face higher import bills and may have to expand subsidies or accept a loss of purchasing power. The IMF says countries benefiting from favorable terms of trade or the AI investment cycle are performing relatively better, widening the gap between economies rather than lifting all boats equally. (IMF Annual Meetings briefing)
AI investment provides the counterforce. Spending on data centers, semiconductors, software and electricity infrastructure is supporting demand and boosting trade. IMF officials describe it as a positive demand shock, in contrast with the negative supply shock caused by energy disruption. The technology boom is therefore helping prevent a sharper global slowdown, at least while investment continues. (IMF press briefing)
The concentration problem
The apparent strength of the AI cycle conceals an important vulnerability: its benefits and supply chains are concentrated among a small number of firms, sectors and countries. Advanced chips, cloud capacity, energy infrastructure and specialized talent are not distributed evenly. A disruption affecting one major supplier, shipping route or source of financing could therefore have effects disproportionate to its immediate location.
Financial commitments are also expanding rapidly. Reuters Breakingviews reported estimates that AI-related capital expenditure could reach $7.6 trillion between 2026 and 2031, with major technology companies potentially issuing $420 billion of debt in 2027 alone. That scale could produce lasting productivity gains, but it also raises the cost of disappointment if commercial returns fail to match expectations. (Reuters Breakingviews)
Optimists argue that today’s infrastructure spending will eventually translate into cheaper services, higher productivity and new industries. The IMF also identifies stronger AI investment and earlier productivity gains as upside scenarios. Skeptics counter that markets may be pricing in benefits before companies can demonstrate durable revenue, leaving investors and heavily indebted firms exposed to a correction.
What comes next
The immediate test is whether energy prices remain elevated. A diplomatic breakthrough could ease inflation and restore confidence, while a wider regional conflict would intensify pressure on central banks and public finances. Even without escalation, prolonged uncertainty may delay investment outside the technology sector.
Interest rates will be central. Higher yields raise borrowing costs for governments and companies just as fiscal buffers are thinning. The IMF has warned that debt-service burdens could crowd out spending on infrastructure, health and climate adaptation. Central banks must therefore avoid allowing temporary energy inflation to become embedded, without aggravating a slowdown caused by weaker real incomes.
Policy choices will determine whether the AI boom becomes a broad productivity cycle or a narrow financial surge. Governments can reduce vulnerability by diversifying chip and energy supply chains, improving electricity networks, expanding worker training and maintaining competition in digital markets. They must also preserve enough fiscal capacity to support exposed households without locking in inefficient subsidies.
The central conclusion is neither that AI will rescue the world economy nor that the boom is bound to collapse. Growth can remain resilient while risks accumulate underneath it. The next phase will show whether technology’s productivity gains spread widely enough to offset geopolitical fragmentation—or whether concentrated investment, expensive energy and high public debt turn resilience into a temporary illusion.