For the better part of two years, Wall Street has treated AI as the most bullish trade on the board, a growth engine that turbocharges earnings, underwrites stretched valuations, and promises a productivity windfall somewhere down the road.
However, the Fed has access to the same numbers and seems to be more inclined to treat the AI build-out as a fresh source of demand in a market that’s still fighting to drag inflation back toward its 2% target.
Goldman Sachs now expects AI-related capital spending to approach $800 billion in 2026, and it calculates that the surge will lift its full-year business investment forecast to 7.8% while adding roughly 3.3 percentage points to capital-expenditure growth on its own.
TrendForce, tracking the nine largest cloud providers in the world, places their combined 2026 outlay near $830 billion, a jump of about 79% over the previous year. A pretty big slice of that increase reflects rising prices rather than added capacity, with Microsoft attributing some $25 billion of its $190 billion budget to costlier memory and components.
All of it puts quite a bit of weight on the inputs the Fed tends to watch most closely, which could turn this investment boom into a policy headache.
Where does the $800 billion in AI spending actually go?
It helps to imagine this spending in physical terms. All of that money takes the shape of land, steel, transformers, copper wiring, gigawatts of fresh generation capacity, industrial-scale cooling, and the incredibly skilled and incredibly rare trades hired to assemble all of it.
Goldman described this as a wave that reaches across servers, semiconductors, memory, power infrastructure, data centers, software, and research budgets, and the bank’s longer-range model traces annual AI capex climbing from around $765 billion this year toward $1.6 trillion by 2031.
Power has become the binding constraint. In a late-May speech, Fed Governor Lisa Cook noted that electricity and water prices have each climbed about 5% over the past year, that chips, high-tech equipment, and software have all grown more expensive, and that wages in specialty construction trades have picked up notably. Households feel some of that pressure on their monthly bills, which began drawing political pushback as several state legislatures move to slow large data-center development.
The central bank’s leadership has been unusually clear and honest about where this leads. Speaking back in March, Jerome Powell told reporters that the construction frenzy was “putting pressure on all kinds of goods and services that go into building these things,” and he conceded that the effect was “probably pushing inflation up.”
Cook went further in that same May address, warning that “yet another shock to prices could be layered on from the heightened investment demand due to AI” and pointing out that companies have announced more than $1.5 trillion in data-center plans, only a sliver of which has actually been built.
The demand side of AI, in other words, is showing up in the price data well ahead of any productivity payoff the technology eventually delivers.
What it means for Bitcoin’s rate-cut bet
The consequences travel from Silicon Valley balance sheets straight into crypto. Bitcoin spent most of the year leaning on the expectation that cooling inflation would free the Fed to cut rates, loosen financial conditions, and rekindle the risk appetite that powered the 2024 rally.
CryptoSlate has documented how tightly the asset now tracks liquidity cycles, a sensitivity that has overtaken Bitcoin halving as the dominant price driver. An $800 billion demand makes rate cuts unlikely, since every dollar of AI-related price pressure hands the Fed one more reason to stay put.
Markets have already begun repricing that. Futures and prediction markets now put the odds of a hold at the June 16-17 meeting above 93%, which will be the first one chaired by Kevin Warsh following his May handover from Powell. CryptoSlate has tracked the reversal as it unfolded, from a stretch when bond traders were pricing a year-end hike to the inflation prints that kept the Fed frozen.
The repricing has bled into spot prices, with Bitcoin sliding to around $63,600 by June 4 after briefly breaking below $62,000, roughly half its October 2025 record and down more than 13% over the week. Much of that damage comes from exits, since Bitcoin ETFs saw a record 11-session outflow streak worth about $3.45 billion, the longest run of redemptions since the funds launched in 2024. A large share of that capital rotated straight into the AI and semiconductor equities that were driving the macro problem in the first place.
Over a five-year horizon, AI may well do what its champions promise, lowering costs, automating routine labor, and easing inflation through real gains in output per worker. However, the build-out phase tends to work the other way around first. Pulling years of infrastructure demand into a narrow window bids up hardware, energy, and talent long before we see any real efficiency, so the price shock arrives early and the windfall arrives late.
That gap between immediate consequences and delayed benefits is what’s been troubling the Fed. Warsh has argued that AI will prove “structurally disinflationary” and usher in “the most productivity-enhancing wave of our lifetimes,” a view that confirms his openness to lower rates. But Cook and Governor Michael Barr lean the other way, with Barr saying flatly that he doesn’t believe the AI boom will be a reason for lowering policy rates.
Traders, on the other hand, have been mostly troubled by timing. Bitcoin, alongside equities and the rest of the market, tends to respond to the first decision in front of them. So, a “productivity thesis” that will probably pay off in 2030 does little to positions held this week, month, or even quarter. Inflation running above 3% leaves Warsh little room to act on his convictions in June, regardless of where he’d like to steer.
The same AI boom inflating tech valuations and carrying the indices higher may be the very force keeping the Fed cautious, delaying the liquidity cycle that crypto traders have spent eighteen months waiting for. If policymakers settle on seeing $800 billion in annual spending as one more pillar of sticky demand, Bitcoin’s rate-cut trade rests on a foundation considerably thinner than its holders would care to admit.
The post AI’s $800 billion spending boom is becoming Bitcoin’s Fed problem appeared first on CryptoSlate.
