AI’s Trillion-Dollar Build-Out Is Complicating the Fed’s Inflation Fight
Silicon Valley leaders from Elon Musk to OpenAI CEO Sam Altman have hyped the deflationary effects of the artificial intelligence boom. “Intelligence too cheap to meter is well within grasp,” Altman wrote recently. Musk has argued that AI and robotics will create extreme abundance and drive down costs. SoftBank’s Masayoshi Son said he expected a 40% drop in prices and that “unnecessarily hard work, sweating work, would no longer be needed.”
None of those dreams are close to being realized. Instead, AI is hitting a wall of corporate inertia as it spreads into the economy — causing near-term inflation and producing little evidence of a sustained productivity boom.
The Spending Numbers Are Staggering
Capital expenditure on the AI build-out is expected to reach $581 billion this year in the U.S., and as much as $1 trillion globally, according to Goldman Sachs Research. U.S. spending alone amounts to 1.8% of GDP — a share expected to rise to 2.8% by 2028.
This spending spree has snarled supply chains and is raising prices in sectors like electricity, chips, software, and data center capacity. Costs are piling up before the full-scale payoff arrives, posing a dilemma for the Federal Reserve under new Chair Kevin Warsh.
Corporate Adoption Is Slower Than Promised
A Census Bureau survey published in May found that between 17% and 20% of U.S. businesses reported using AI — far more prevalent at large firms than small ones. Corporate adoption remains uneven, delaying the highly anticipated productivity gains that could offset the infrastructure costs.
Ronnie Chatterji, chief economist at OpenAI, acknowledged the gap: “For it to impact the economy, it has to be adopted by organizations. Those organizations have to realize value. It’ll still be a little while before we see it sort of clearly for productivity statistics.”
The ‘Weak Links’ Problem
Julie Averill, Lululemon’s former chief information officer who oversaw AI adoption at the company, offered a grounded perspective: “The reality is that the technology is there. The hype is around the ease of the technology in a large organization.”
She explained that the challenges of implementing AI in large companies remain the same as always: people. “Getting people to change their behaviors, taking them along the journey with you, and getting them to trust the model — that’s hard.”
Economists use the term “weak links” for tasks that can’t be easily automated. AI makes us more productive by automating work like reading a radiological scan — something AI can do very well. But jobs are bundles of tasks. The Nobel laureate Geoffrey Hinton predicted in 2016 that radiologists wouldn’t be needed within five to 10 years. Instead, their numbers kept growing as AI made radiologists more valuable to the economy.
“It turns out that radiologists do more than just read scans,” explained Stanford professor Charles Jones, a leading scholar of how AI affects growth, now on leave at Anthropic. “AI tools complement those other skills by automating a fraction of the tasks that radiologists perform.” The other things radiologists do — talking to patients, working with colleagues — fall in the category of weak links.
The Power User Gap
Chatterji revealed data from OpenAI showing that AI power users deploy the technology at eight times the rate of average companies, measured by tokens per user. The gap has grown from two times since OpenAI published a report on it just three months ago.
“It is growing incredibly fast in terms of the gap between the frontier firms and the typical firms. The companies that are reorganizing their workflows around it and changing the way they work around AI, they’re having more success.”
Ronnie Chatterji, Chief Economist at OpenAI
Silicon Valley Advises the Fed
Fed Chair Kevin Warsh last month appointed Jones to a task force that will inform how the central bank thinks about AI and its effect on the economy. Venture capitalist Marc Andreessen, whose firm is aggressively backing AI startups, is also on the team and is among those predicting an era of “hyper-deflation.”
Peter Boockvar, chief investment officer of OnePoint BFG Wealth Partners, offered a skeptical counterpoint: Even during the internet-driven productivity boom, the U.S. saw only a 1.5% gain in productivity over 30 years. “To think that generative AI is going to bring that level of enhancement to the economy, relative to the internet, is tough. Technology has always made people more productive. But is generative AI multiple step functions higher? We just don’t know.”
Key Takeaways
- AI infrastructure spending: $581B in the U.S. (1.8% of GDP), ~$1T globally in 2026
- Only 17-20% of U.S. businesses use AI, with adoption heavily skewed toward large firms
- The “weak links” problem: AI automates some tasks well but jobs are bundles of tasks
- AI power users deploy tech at 8x the rate of average companies — and that gap is widening
- Fed Chair Warsh formed an AI task force including Stanford’s Charles Jones and VC Marc Andreessen
- The mismatch between AI costs (inflationary) and benefits (deflationary) creates a policy dilemma
Source: CNBC — Matt Peterson & Kate Rooney, August 12, 2026
