BlackRock Sees AI Infrastructure Spend Accelerating

Global investment in AI intelligence infrastructure shows no signs of slowing, according to BlackRock’s Ben Powell.
Powell, chief investment strategist for APAC at BlackRock, argues that the companies supplying the backbone of the AI ecosystem remain the most reliable winners in the current technology cycle.
He didn’t pen these thoughts on the back of a dollar bill, but in fact spoke with CNBC today (Dec. 8) at Abu Dhabi Finance Week. Powell reckons the rapid expansion of data centers, chip manufacturing, and power generation is creating what he called a “traditional picks and shovels capex super boom.”
“The capex deluge continues. The money is very, very clear,” Powell told CNBC, adding that BlackRock sees the infrastructure layer of AI — from chips and servers to electricity providers and copper-wire manufacturers — as poised for longer-lasting gains than the AI model developers receiving most public attention. “It still feels like it’s got more to go,” he said.
His comments come during a year in which AI capital expenditure has become one of the primary forces driving global markets. Large technology companies continue to pour unprecedented sums into hardware, cloud expansion, and power procurement as they race for dominance in what they increasingly view as a winner-takes-all contest.
Hello to new highs
Nvidia has been the most visible beneficiary of that spending surge. Its GPU chips, essential to training and running modern AI models, propelled the company to briefly surpass the $5 trillion market-cap milestone this year — a moment that intensified debate over whether markets are entering an AI bubble.
Meanwhile, Microsoft and OpenAI formalized a restructuring agreement in October designed to support OpenAI’s capital-raising needs. Reuters has reported that OpenAI is exploring an initial public offering that could value the company at $1 trillion.
Alongside high-profile model developers, another wave of investment is reshaping supply chains behind the scenes. Chip supply agreements are being secured years in advance, power contracts are expanding across continents, and grid operators from the US to the Middle East are scrambling to meet surging energy requirements for new data centers. Amazon, Meta, and other hyperscale firms have each budgeted tens of billions annually to support AI-driven infrastructure development.
S&P Global projects that data-center electricity demand could nearly double by 2030, a shift with major implications for energy markets, climate targets, and regional power planning. Growth is expected to come not only from hyperscalers but also enterprise facilities, leased co-location centers, and even crypto-mining operations competing for the same limited power resources.
Next investment phase
Powell says these firms have only just begun to explore debt financing as a tool for expanding AI-related operations. “The big companies have only just started dipping their toes into the credit markets… feels like there’s a lot more they can do there,” he said. This suggests that a new wave of borrowing — and therefore investment — may be on the horizon, especially as companies seek to secure everything from chip supply to long-duration energy contracts.
Executives, Powell noted, are operating under the belief that coming second in AI could mean being shut out of the market entirely. That mindset is prompting aggressive outlays, even at the risk of overshooting or investing ahead of proven demand. The result is a global acceleration in spending on components and utilities rather than on the consumer-facing AI applications currently capturing public attention.
Infrastructure may outperform AI developers
Powell’s view highlights a growing consensus among institutional investors: while AI software companies may see sharper volatility, the hardware, materials, and energy sectors supporting them could offer steadier returns. “If we’re the recipients of that cash flow, I guess that’s a pretty good place to be, whether you’re making chips, whether you’re making energy all the way down to the copper wiring,” he said. Powell expects “positive surprises driving those stocks in the year ahead.”
This perspective carries meaningful implications for markets. It suggests that the sustainability of the AI boom may depend less on model breakthroughs and more on the ability to build, power, and maintain the physical systems behind them. It also underscores how the race among hyperscalers is reshaping global capital flows, accelerating grid upgrades, and transforming industrial sectors far beyond Silicon Valley.
As AI adoption expands, the infrastructure layer appears increasingly central — not only to technology firms’ competitive strategies but also to global financial markets, commodity producers, and national energy planners.
Meta is redirecting its financial efforts to AI, but this time the shift feels less like chasing a futuristic dream and more like a course correction in response to reality.
The post BlackRock Sees AI Infrastructure Spend Accelerating appeared first on eWEEK.