🔍 Read the full analysis: What We Know About AI’s Payoff Beyond The Tech Industry on ThorstenMeyerAI.com
Get business pricing on office and shipping supplies
- Business-only prices and quantity discounts
- Tax-exempt purchasing
- Multiple users, one account, clear invoices
TL;DR
Apollo economist Torsten Slok says recent productivity, revenue-per-worker and margin data show no clear financial payoff from AI across companies outside the technology giants. That is an interpretation of broad trends, not proof that AI has produced no gains: controlled workplace studies report improvements in specific tasks, while most executives surveyed say they have not yet seen a productivity effect.
Apollo chief economist Torsten Slok says recent data show no clear financial payoff from AI across companies outside the major technology firms, even as investment in computing capacity climbs. His October 24 analysis points to flat revenue per worker and profit margins among the other 493 S&P 500 companies, but it measures broad economic patterns rather than isolating AI’s effect on each business.
Apollo compared inflation-adjusted revenue per worker and profit margins for the Magnificent Seven—Alphabet, Amazon, Apple, Meta, Microsoft, Nvidia and Tesla—with the other 493 companies in the S&P 500. Its charts show revenue per worker rising for the seven and remaining broadly flat for the rest, while the seven’s margins climbed to about 30% and the other group’s stayed near 10%. Apollo also cited San Francisco Federal Reserve productivity measures that, in its reading, show no clear AI-related acceleration.
The analysis draws on figures from Ramp, a corporate spending platform whose data cover more than 85,000 U.S. businesses. Ramp data indicate that the top 10% of firms account for 99.5% of spending on model serving and inference, 99.0% of neocloud spending and 95.6% of spending on OpenAI and Anthropic. Those figures describe Ramp’s customers, not the S&P 500 or all U.S. businesses. Apollo also cited a survey of nearly 9,000 executives in which 83% reported no productivity impact from AI so far; the source material gives an average productivity effect of 0.2% a year.
Meanwhile, company filings show the scale of investment commitments. Meta reported $34.91 billion in capital expenditures in its third quarter and narrowed its 2026 capital-spending outlook to $135 billion to $145 billion. Its filing lists about $355.73 billion in leases that have not yet started and $454.52 billion in non-cancelable contractual commitments, mostly for cloud capacity, servers and network equipment, plus up to $14.72 billion in contingent cloud purchases. Together, those items total about $825 billion; they are future commitments, not money already spent, and not all relate to AI.
Infographic 01 · The “zero” in numbers
What “zero financial payoff” actually measures.
Apollo’s headline rests on five charts. Here are their numbers, and what they can and cannot prove.
Who pays for AI? Share of spending by the top 10% of firms
Ramp customer data via Apollo · non-AI categories in grey for comparison
Spending is concentrated in every category, not only AI; AI is the most extreme.
What the five charts do and do not show
shown not shown
- Shown: revenue per worker and margins rising for the 7 tech giants, flat for the other 493
- Shown: no AI lift yet in US productivity statistics (San Francisco Fed TFP, output per hour)
- Shown: AI spending sits in few, mostly venture-funded firms (Ramp, 85,000+ US businesses)
- Not shown: that AI caused the gap; rates, tariffs and wages move the same lines
- Not shown: gains inside single functions, which can cancel out in an average
- Not shown: the S&P 493 itself in the Ramp data; those are Ramp’s customers
“The AI story will be decided in the S&P 493, not the Mag 7.”Torsten Slok, Apollo chief economist, 24 Oct 2026 (opinion)
Sources: Apollo Daily Spark (24 Oct 2026) with charts by Bloomberg, Macrobond, SF Fed, BLS and Ramp; Apollo (20 Oct 2026); Ramp AI Index. Values with ≈ are read off Apollo’s charts. Assessment: Thorsten Meyer.
Powered by Thorsten Meyer AI
Where AI Gains Are Showing Up
The difference between broad corporate indicators and task-level results matters to businesses weighing whether to expand AI use. Aggregate figures can show that widespread gains are not yet apparent, but they cannot tell a company whether a particular tool saves time, lifts sales or improves service in a particular workflow. A flat average may also conceal companies that gained and others that did not.
The source material cites customer-support research showing 14% more issues resolved per hour, rising to 34% for novice workers, and retail experiments with sales effects ranging from no increase to 16.3%, depending on the workflow. These are specific results, not evidence that the same gains apply across industries or companies. They suggest that AI’s payoff may depend on how a system is introduced, what task it handles and whether workers can use it effectively.
For readers, the gap between large investment plans and uncertain returns has consequences beyond technology companies. Spending on data centers, chips and cloud services can affect suppliers, lenders and the cost structure of firms adopting AI. If revenue or productivity gains fail to keep pace with that spending, companies may face pressure to show results, reduce costs or reconsider expansion plans. The evidence available here does not establish that those consequences will occur, but it explains why investors and executives are scrutinizing returns.
Popular right now
Philips 4400 Series Fully Automatic Espresso Machine with LatteGo Milk System
Breville Bambino Espresso Machine, Brushed Stainless Steel
Dyson V8 Cyclone Cordless Vacuum
As an affiliate, we earn on qualifying purchases.
The Spending Behind the Debate
Apollo’s note, titled “No Signs of AI Payoff Outside the Tech Sector,” followed a period of major investment announcements and financial disclosures. Its central comparison is between the seven large technology companies, which are direct sellers of technology and major AI investors, and the remaining S&P 500 companies. The contrast describes how the groups’ reported revenue per worker and margins have moved; it does not prove AI caused the difference.
Meta’s third-quarter results provide one example of the spending involved. The company reported $64.60 billion in revenue, up 26% from a year earlier, and $32.79 billion in operating cash flow. With capital expenditures of $34.91 billion, free cash flow was negative $2.12 billion, compared with positive $10.6 billion a year earlier. Those figures cover Meta’s overall business, so they do not show how much of its cash flow change came from AI specifically.
The $825 billion commitment figure combines different kinds of obligations, including leases for data centers and energy infrastructure that run through 2036, and contracts for cloud capacity and equipment. It should not be read as a single-year bill or as an AI-only total. Forecasts cited for the five largest technology spenders—Amazon, Alphabet, Meta, Microsoft and Oracle—are also projections, not company commitments. The distinction between realized spending, contractual obligations and forecasts is central to judging the size of the investment cycle.
“No Signs of AI Payoff Outside the Tech Sector”
— Torsten Slok, Apollo chief economist, in the October 24 Apollo Daily Spark note
What Aggregate Data Cannot Show
Apollo’s “zero” framing is stronger than the available measurements can establish. The cited revenue, margin and productivity trends do not isolate AI from other influences, including interest rates, tariffs and labor costs. Nor can a group average identify companies whose gains may be offset by losses elsewhere. The evidence supports a narrower conclusion: a broad financial payoff is not yet visible in these measures.
The source material does not provide enough detail to assess the methods, sample sizes or time periods behind every task-level result. The Ramp spending figures cover its own customers, while the executive survey records reported effects rather than independently measured productivity. It is also unclear how much of Meta’s future commitments supports AI rather than other operations, including Reality Labs hardware.
More broadly, the materials do not establish whether current investment will produce later gains, how those gains might be distributed, or how much spending will translate into revenue. Meta’s negative free cash flow in the quarter is a company-wide result and cannot by itself be attributed to AI. These limits mean the data describe an unsettled payoff, not a final verdict on the technology.
Evidence to Watch in Coming Quarters
The clearest next test will be whether companies report measurable productivity, revenue or margin improvements as AI systems move from trials into routine work. Investors will also watch how major technology companies fund data centers and computing capacity, and whether their spending plans change as results come in. Reported commitments and forecasts should be distinguished from quarterly cash expenditure.
For Meta, subsequent earnings reports can show how capital spending, cash flow and contractual obligations develop against its stated 2026 outlook. Across sectors, better evidence will require measures that compare results before and after AI adoption while accounting for other business changes. Until those data accumulate, the broad claim that AI has paid off—or has not paid off—across corporate America remains too sweeping.
Key Questions
What did Apollo mean by saying AI has no payoff outside tech?
Apollo’s analysis found that revenue per worker and profit margins rose for seven large technology companies while remaining broadly flat for the other 493 S&P 500 firms. That is an aggregate comparison, not proof that no non-tech company or task has benefited.
Does the $825 billion figure mean Meta has already spent that amount?
No. It is a combined estimate of future leases and contractual commitments, plus contingent cloud purchases. The obligations span multiple years, and the source material says not all of them relate to AI.
Have workplace studies found AI productivity gains?
The source material cites task-level results, including 14% more customer-support issues resolved per hour and larger gains for novice workers. It also cites retail experiments with sales effects from 0% to 16.3%, depending on the workflow. These findings do not establish a general result for all workplaces.
Why can the overall data show little impact if some tasks improve?
Task-level gains may not yet be widespread, may not translate into higher company revenue or margins, or may be offset by costs and weaker results elsewhere. Group averages can also conceal differences between companies. The available figures do not determine which explanation is most important.
Source: ThorstenMeyerAI.com
