The global economy is absorbing the latest Middle East energy shock better than initially feared, and a major reason is the unusually strong investment cycle surrounding artificial intelligence. Spending on data centres, advanced computing equipment and semiconductor production is supporting economic activity at a time when higher energy prices are weakening household purchasing power and raising costs for businesses.

The Organisation for Economic Co-operation and Development expects global growth to reach 2.9 percent in 2026, slightly above its earlier projection, before rising to 3 percent in 2027. The modest improvement in the current-year forecast reflects the strength of private investment and technology-related production, particularly in economies closely connected to the artificial intelligence supply chain. However, the outlook for 2027 has been reduced because the energy shock is proving more persistent than previously expected.

The contrast is important. Artificial intelligence is creating a powerful source of demand precisely when the energy crisis is creating a powerful source of pressure. Data centre construction requires large amounts of equipment, engineering, electricity and advanced chips, while semiconductor manufacturers and technology exporters benefit from the resulting investment cycle. This spending is generating economic activity well beyond the companies directly developing artificial intelligence.

Yet the same technology boom that is supporting growth also has an important vulnerability. Data centres are highly dependent on electricity, and shortages or sharply higher energy prices could eventually increase the cost of expanding artificial intelligence infrastructure. The current resilience therefore reflects a temporary balance between two opposing forces rather than the disappearance of the energy shock.

AI Spending Is Supporting Growth Beyond Technology

The economic importance of artificial intelligence investment comes from the scale and breadth of the infrastructure required to develop and operate advanced systems. Data centres need specialised processors, servers, networking equipment, cooling systems, construction services and electricity. Semiconductor factories require enormous capital expenditure, while supporting industries provide additional demand for machinery, engineering and construction.

This creates a multiplier effect. A company building a data centre is not simply purchasing computer equipment from one technology supplier. It may require construction contractors, electrical equipment, cooling systems, power infrastructure and a large network of semiconductor and component suppliers. Strong demand in one part of the technology sector can therefore spread into manufacturing and services.

The United States has been one of the clearest beneficiaries. Heavy investment in data centres and technology equipment has helped offset weaker consumer spending and slowing real household income growth. The OECD expects the United States to grow by 2.2 percent in 2026 and 2.1 percent in 2027, with artificial intelligence investment providing an important source of momentum.

The effect extends into Asia. Japan and South Korea are major suppliers of semiconductor and technology components, allowing their economies to benefit from the investment taking place elsewhere. China is also benefiting from technology-related production, while other emerging economies are receiving support from domestic demand and government measures designed to cushion higher energy costs.

This explains why the global economy has not reacted to the energy shock in the same way as it might have under different conditions. Artificial intelligence investment has created an additional source of industrial demand that was not present during previous energy disruptions.

The Energy Shock Is Still Eroding Purchasing Power

The strength of AI spending should not obscure the damage caused by higher energy prices. Oil and gas are inputs into transportation, manufacturing, agriculture and household consumption, meaning that an extended energy shock eventually reaches most parts of the economy.

The first effect is straightforward: households pay more for fuel, heating and electricity. That leaves less money available for discretionary spending, putting pressure on consumer-facing businesses. Companies face a similar problem because higher energy and transportation costs reduce margins unless those costs can be passed on to customers.

The OECD expects this pressure to become more visible over time. Household savings and corporate inventories can absorb higher costs temporarily, but those buffers are not unlimited. As they are gradually depleted, the impact of higher energy prices can become more visible in consumption and investment.

This creates a different economic environment from the early stages of the shock. Initially, companies may rely on inventories, alternative suppliers and existing cash reserves. Governments can also provide temporary support. But if energy prices remain elevated for an extended period, these mechanisms become increasingly expensive or ineffective.

The OECD has therefore warned that the strongest drag on global growth could emerge around the end of 2026 and the beginning of 2027. Inflation is also expected to remain higher for longer because energy costs are being transmitted through the wider economy.

AI Itself Is Becoming Exposed to Energy Costs

One of the most important complications is that artificial intelligence is not independent of the energy market. The infrastructure supporting advanced AI systems consumes substantial amounts of electricity, particularly as companies build larger data centres and deploy increasingly demanding computing systems.

This creates a paradox. AI investment is helping the global economy absorb the energy shock, but an extended energy shock can also make AI investment more expensive. Higher electricity prices increase data centre operating costs, while shortages of available power can delay new facilities.

The OECD has specifically identified electricity availability as a potential constraint on future AI investment. If energy infrastructure cannot expand quickly enough to meet demand from data centres, the cost of deploying AI systems could rise and some planned investments could be delayed.

This means that AI cannot indefinitely compensate for an energy shortage simply by attracting more investment. The technology sector itself requires reliable and affordable power. Data centres also compete with households, manufacturers and other industries for electricity, creating another potential source of pressure.

The relationship between energy and AI is therefore becoming increasingly important. The current investment boom is sustainable only if the physical infrastructure supporting it expands alongside demand.

The Biggest Risk Is That AI Investment Disappoints

Another vulnerability lies in the financial expectations surrounding artificial intelligence. Companies are investing enormous sums based on the assumption that AI will eventually generate significant productivity improvements and new sources of revenue. If those returns take longer to appear than expected, investment could slow sharply.

That would matter because the current AI investment cycle has become large enough to influence economic growth, financial markets and international trade. A substantial reduction in technology investment would affect data centre construction, semiconductor production, equipment manufacturers and other suppliers.

The OECD has warned that disappointing returns from AI investment could trigger a reassessment of corporate valuations and reduce spending. This risk is particularly important because the current investment boom depends partly on expectations about future demand rather than only on existing revenues.

There is already a distinction between AI adoption and AI monetisation. Companies are rapidly increasing spending on computing infrastructure, but the economic benefits of that spending will depend on whether businesses can eventually use AI to increase productivity, reduce costs or generate new revenue.

If those benefits emerge gradually, investment can continue while the technology becomes integrated into ordinary business operations. If returns fail to meet expectations, the adjustment could be abrupt.

Why the Current Resilience May Not Last Indefinitely

The OECD’s outlook therefore presents two different timelines. In the near term, artificial intelligence investment is helping offset weaker consumer demand and the effects of higher energy prices. In the longer term, the persistence of the energy shock could reduce household purchasing power, increase inflation and raise the cost of the very infrastructure driving the AI boom.

Global growth is projected at 2.9 percent in 2026 and 3 percent in 2027, but those numbers depend on energy prices eventually easing and AI-related activity remaining strong. The outlook would become considerably weaker if the Middle East conflict caused prolonged disruption to oil and gas supplies.

A prolonged energy shock could also create shortages of specialised inputs and raise costs for industries that depend heavily on electricity. Artificial intelligence would not be immune. Data centres require continuous power, and advanced semiconductor production depends on complex industrial infrastructure that itself consumes significant energy.

Central banks face another complication. Higher energy prices raise inflation while weaker real incomes reduce demand. If inflation spreads into wages and services, monetary authorities may need to keep interest rates higher for longer. That would increase financing costs for companies investing in data centres and other AI infrastructure.

The current global economy is therefore being supported by a powerful but increasingly complicated combination of forces. Artificial intelligence investment is generating manufacturing activity, technology exports, construction and capital spending that help compensate for weaker consumption. At the same time, the energy shock is steadily reducing purchasing power and raising operating costs across the economy.

The important point is that AI is acting as a cushion, not a solution. Its investment boom can soften the immediate economic impact of higher energy prices, but it cannot permanently offset an energy shortage. The technology industry itself depends on reliable power, affordable financing and strong demand.

For now, the balance remains favourable enough for global growth to hold up better than expected. But the longer the energy shock persists, the more difficult it becomes for AI investment alone to compensate for its effects. The sustainability of the current growth pattern will ultimately depend on whether energy markets stabilise before the costs of the disruption begin to overwhelm the investment momentum generated by artificial intelligence.

(Adapted from Binance.com)

Categories: Economy & Finance, Geopolitics, Regulations & Legal, Strategy