Why AI’s Payoff Is Mostly Staying Within The Tech Sector

🔍 Read the full analysis: Why AI’s Payoff Is Mostly Staying Within The Tech Sector on ThorstenMeyerAI.com

Before you orderOffer from Amazon

Get smart everyday buys delivered free with Prime

• Fast, free delivery on millions of items
• Prime Video, Amazon Music and more included
• Member-only deals all year

Start your free Prime trial →Free trial for eligible customers · Cancel anytime

As an affiliate, we earn on qualifying purchases.Hide

Apollo Global Management chief economist Torsten Slok says recent data show no visible financial payoff from artificial intelligence across corporate America outside the technology sector, even as companies pour money into computing infrastructure. His October 24 note compares productivity and financial trends among major technology firms with those of other S&P 500 companies; it does not establish that AI has delivered no benefit at any individual non-tech company.

Apollo’s charts show inflation-adjusted revenue per worker rising for the “Magnificent Seven” — Alphabet, Amazon, Apple, Meta, Microsoft, Nvidia and Tesla — while the other 493 companies in the S&P 500 have remained roughly flat. The note also shows profit margins for the seven rising to about 30%, compared with roughly 10% for the other companies. These comparisons describe observed corporate trends; they do not isolate AI as the cause.

The note cites San Francisco Federal Reserve productivity measures as showing no signs of AI boosting productivity growth so far. It also uses spending data from Ramp, a corporate card and spending platform with more than 85,000 U.S. business customers. In those data, the top 10% of firms account for 99.5% of model-serving and inference spending, 99.0% of neocloud spending and 95.6% of spending on OpenAI and Anthropic. Those figures describe Ramp customers, not every U.S. business.

The gap between spending and broad measurable returns is also visible in company filings. Meta reported $34.91 billion in capital expenditures in the third quarter and narrowed its 2026 spending forecast to $135 billion–$145 billion. Its filing listed about $355.73 billion in leases not yet started and $454.52 billion in non-cancelable contractual commitments, plus up to $14.72 billion in contingent cloud purchases. The combined figure is roughly $825 billion in future commitments; it includes items beyond AI, including Reality Labs hardware, and is not the same as spending already paid.

The Divide Between AI Costs and Returns

The findings sharpen a question facing investors, business leaders and workers: who is capturing the economic value of AI, and when will it spread beyond the firms building and supplying the technology? The largest technology companies are reporting stronger revenue per worker and margins while making large infrastructure commitments. For many other companies, Apollo’s aggregate measures do not yet show comparable gains.

That does not mean AI tools have no practical value. It means the financial impact is difficult to see in economy-wide or company-wide measures so far. If spending grows faster than revenue or productivity, AI infrastructure could weigh on margins. If targeted deployments improve specific tasks and those improvements scale, they may later appear in broader results. The current evidence leaves both possibilities open.

Financing adds another dimension. The source material reports that SpaceX borrowed $40 billion to buy AI chips and describes rising concern about the cost of funding infrastructure providers. Large investment plans can support future capacity, but they also create exposure if demand or returns fall short. The scale of commitments makes the distinction between investment plans, contractual obligations and realized returns especially relevant.

What Apollo’s Comparisons Measure

Apollo’s October 24 note, titled “No Signs of AI Payoff Outside the Tech Sector,” follows a debate about whether the rapid build-out of AI infrastructure is translating into gains for businesses using the tools. The note’s “zero” wording is Slok’s interpretation of aggregate indicators, not a direct measurement showing that every company outside tech has received no benefit. Revenue per worker and profit margins can shift for many reasons, including labor costs, demand, interest rates and industry mix.

The source material also cites specific workplace studies that point to gains in some settings. One customer-support study reported 14% more issues resolved per hour, with a 34% increase for novice workers. Retail experiments reported sales effects ranging from zero to 16.3%, depending on the workflow. These findings concern particular applications, not the average performance of the wider economy.

At the same time, a survey of nearly 9,000 executives found that 83% reported no productivity impact so far, while the average reported impact was 0.2% a year. The survey result and the task-level studies measure different things: one asks about reported effects across firms, while the others track outcomes in selected workflows. Together, they help explain how isolated improvements can coexist with little change in broad productivity statistics.

““Artificial intelligence has delivered zero financial payoff to corporate America outside of tech.””

— Torsten Slok, Apollo chief economist, in the October 24 Daily Spark note

How Much AI Drives the Gap

The data do not establish causation. Rising revenue per worker or margins at the largest technology firms cannot by themselves show that AI produced the improvement, and flat averages among the other 493 companies can conceal businesses that gained or lost. Apollo’s aggregate comparisons also cannot separate AI’s influence from other economic forces.

The spending figures have limits too. Ramp’s customer data are not a census of businesses, Meta’s commitments include non-AI activity, and projected industry spending is not equivalent to money already spent or returns already earned. The source material does not provide a common, company-by-company measure of AI investment and financial benefit across sectors.

It is also unclear when task-level improvements will translate into sustained productivity growth or higher profits across entire companies. The available examples vary by workflow, and the cited executive survey records little reported productivity impact to date. The size and timing of any broader payoff remain unsettled.

Evidence to Watch in Results

Investors and companies will be able to test the competing interpretations against future financial reports and productivity data. Useful signals include whether firms outside the major technology companies report sustained gains in output per worker, stronger margins or lower costs tied to specific AI deployments, rather than simply announcing new tools or infrastructure budgets.

For infrastructure providers, the next test is whether customer demand supports the capital commitments and borrowing described in the source material. Meta’s spending outlook and its future contractual obligations will remain relevant measures of how quickly the build-out is proceeding. Whether large investments produce returns outside the technology sector is not yet known.

Key Questions

What does Apollo mean by “zero financial payoff”?

It is Torsten Slok’s description of aggregate trends: revenue per worker and margins are rising for seven major technology companies, while the other 493 S&P 500 companies are roughly flat in the charts cited. It does not prove that every non-tech company has received no AI-related benefit.

Does the report show that AI has no productivity benefits?

No. Apollo cites broad productivity measures with no visible AI boost so far, but the source material also reports gains in some specific customer-support and retail workflows. Those local results have not yet shown up as broad, economy-wide productivity growth.

What is included in Meta’s roughly $825 billion commitment figure?

The figure combines leases not yet started, non-cancelable contractual commitments and possible contingent cloud purchases reported in Meta’s filing. The commitments are mainly tied to data centers, cloud capacity, servers and network infrastructure, but not all are AI-related, and the total is not cash already spent.

Why are the findings relevant to businesses and investors?

They highlight a widening question about whether large AI investments will produce measurable returns beyond the companies selling technology and computing capacity. Spending, reported productivity gains and financial results need to be tracked separately because they do not measure the same thing.

Source: ThorstenMeyerAI.com