The global economy is being pulled in two opposing directions. Artificial intelligence is driving a historic investment cycle in chips, data centres and cloud infrastructure, while war, trade tensions and higher energy costs are narrowing the room governments and central banks have to absorb another shock.

The tension is visible in the scale of corporate spending. Alphabet, Amazon, Meta and Microsoft have outlined more than $740 billion in combined capital expenditure for 2026, much of it directed toward AI infrastructure, according to reporting cited by The Daily Star. The spending has supported technology suppliers and helped sustain market optimism, but it has also revived an older question: how much investment can be justified before expected future profits become too distant or uncertain?

A boom with an unusually large bill

AI is not simply another software upgrade. The current expansion requires expensive, power-hungry physical infrastructure, including advanced processors, specialised data centres and electricity connections. That gives the boom a wider economic footprint than previous internet cycles, but it also makes returns more dependent on continued demand and on companies’ ability to convert computing capacity into revenue.

Investors are increasingly focused on that conversion. Reuters reported that markets were already strained by wars, sovereign debt and trade disputes as AI spending pushed equities higher and borrowing costs toward multidecade highs. The concern is not that AI has no economic value; it is that prices and investment commitments may be running ahead of measurable productivity gains.

“A downturn has been avoided ‘for now’,” the International Monetary Fund said, while warning that growth remains modest and inflationary pressures persist.

The optimistic case is substantial. AI can raise productivity, support new industries and strengthen countries that lead in semiconductors, cloud services and related exports. The IMF’s July outlook described the technology cycle as a positive shock offsetting part of the negative supply shock from conflict in the Middle East. In that scenario, today’s infrastructure spending becomes the foundation for faster potential growth rather than a speculative excess.

War turns investment into a policy test

The counterforce is geopolitical. The IMF projects global growth of 3.1% in 2026, down from 3.4% the previous year, while identifying wars in the Middle East and Ukraine, trade tensions and commodity disruptions as continuing risks. Higher energy and freight costs spread well beyond conflict zones, raising the prices of transport, food and manufactured goods.

That creates a difficult policy mix. Central banks may need to keep interest rates higher to prevent an energy shock from becoming entrenched inflation. Higher rates, however, increase the cost of financing data centres, public debt and business expansion. They also make long-duration technology valuations more vulnerable if investors demand quicker returns.

Governments face a similar trade-off. Subsidies and industrial policies can attract chip factories, data centres and electricity investment, potentially improving national resilience. But public support can also shift private losses onto taxpayers if demand disappoints. Countries without large technology sectors may instead face higher import bills for energy and computing capacity, widening the gap between AI beneficiaries and the rest of the world.

What comes next

The next phase will be decided less by announcements than by evidence. Markets will watch whether AI companies produce durable cash flow, whether customers continue paying for applications and whether productivity improves outside the technology sector. They will also watch energy availability: a data-centre expansion that cannot secure affordable power will encounter physical limits regardless of investor enthusiasm.

Financial stability is another fault line. The IMF has warned about vulnerabilities in non-bank financial institutions, while Reuters identified the combination of heavy government borrowing, geopolitical volatility and AI investment as a source of wider market risk. A correction in technology shares would not necessarily become a global crisis, but it could weaken business confidence and expose lenders or funds that financed the expansion on optimistic assumptions.

The most plausible outcome is neither an effortless AI-led boom nor an immediate collapse. Investment is likely to continue, but with greater scrutiny, consolidation and pressure to show returns. The countries best positioned will be those that combine technological capability with reliable energy, sustainable public finances and credible regulation.

The larger lesson is that AI cannot be separated from the world around it. Its promise may raise productivity, but its financing depends on stable markets; its infrastructure depends on energy; and its social benefits depend on workers and consumers sharing in the gains. As conflict and debt reduce policymakers’ margin for error, the technology boom is becoming a test of whether innovation can outrun instability.

Sources