# AI-driven growth may run out of steam as energy grids struggle to keep up with innovation > Source: > Published: 2026-07-15 10:20:03+00:00 # AI-driven growth may run out of steam as energy grids struggle to keep up with innovation Data centre pushback in the US signals early limits to the AI capex boom, says StanChart research head [SINGAPORE] The pace of artificial-intelligence innovation risks outrunning the energy and infrastructure capacity needed to support it, said Standard Chartered’s global head of research, Eric Robertsen, on Wednesday (Jul 15). This represents a growing headwind for a global economy that has been buoyed by AI-related capital expenditure. Speaking at the bank’s global research briefing for the second half of 2026, Robertsen said the debate over AI valuations has moved beyond questions of future revenue to a more fundamental capacity constraint. “Not only in the United States but in a number of other economies, there’s simply not enough grid capacity to support all of the energy demand,” he said. Robertsen was referring to New York, which recently became the first US state to impose a moratorium on new AI data centre construction, following a Gallup poll that found more than 70 per cent of respondents opposed such projects in their neighbourhoods. In response, large technology companies have begun building their own power sources without seeking permission, prompting pushback in some communities, he added. This trend suggests capex growth will eventually have to slow, “whether there’s money available for it or not”, said Robertsen, as AI innovation outpaces the grid’s ability to keep up. He also pointed to a shift in the balance sheets of AI-linked companies, which have moved from generating strong free cash flow with little debt to carrying heavy debt loads to fund capex. In the bank’s Jul 9 report, its global research team estimated hyperscalers’ combined planned investments will hit US$650 billion to US$725 billion in 2026, up from US$380 billion to US$410 billion in 2025. The researchers warned that a sharp pullback could hit global growth through both financial markets and real economic activity, given that the “Magnificent Seven” – Apple, Microsoft, Alphabet, Amazon, Nvidia, Meta and Tesla – tech stocks’ market capitalisation is now nearing 20 per cent of global gross domestic product. ## Outsized AI growth in certain markets This issue comes against a backdrop where AI capex spending has become a disproportionately large driver of growth for many economies in 2026. Edward Lee, StanChart’s chief economist, noted that AI-related demand accounted for an estimated 1.35 percentage points of the US’ 1.6 per cent first-quarter GDP growth, with the rest of the economy essentially flat. Similarly, Asian economies plugged into the AI supply chain have seen their growth buoyed by related exports and spending. Lee pointed to Singapore, where AI-related demand accounted for an estimated 40 per cent of the city-state's growth rate in the first half of the year. This concentration makes AI capex spending “more of a risk than a baseline”, said Lee. On the broader global picture, Robertsen said financial markets have grown increasingly numb to geopolitical shocks from the Middle East, even as the underlying economic uncertainty persists. In contrast, China’s economic quiet has been a stabilising factor for the region, and that should not be taken for granted, along with a rising risk of renewed volatility from Japan and the US midterm elections in November. Turning to Asia, Lee said growth prospects remain divided between economies plugged into the AI supply chain, such as Taiwan and South Korea, and those that are not. Divya Devesh, StanChart’s co-head of FX research for Asean and South Asia, added that a resilient US dollar and weak exporter conversion into local currencies continue to weigh on the region’s currencies, though the Singapore dollar and Malaysian ringgit stand out as relative outperformers. On Singapore specifically, Lee said the Monetary Authority of Singapore is likely to hold monetary policy settings steady after tightening in April, with inflation “pervasiveness” – how broadly price increases are spreading across the consumption basket – still running low. That, he said, remains a more important gauge for policy than headline growth figures alone. Decoding Asia newsletter: your guide to navigating Asia in a new global order. Sign up here to get Decoding Asia newsletter. Delivered to your inbox. Free. Copyright SPH Media. All rights reserved.