🔍 Read the full analysis: The Economist’s Take On AI’s Returns And Rising Investment on ThorstenMeyerAI.com
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
TL;DR
Apollo chief economist Torsten Slok says the data show no broad financial payoff from AI for companies outside the technology sector. The claim is based on aggregate productivity, revenue and margin trends, while major companies continue to commit large sums to data centers and computing capacity; specific workplace studies show gains in some tasks.
Apollo chief economist Torsten Slok said October 24 that available data show no broad financial payoff from AI outside the technology sector, even as companies commit heavily to computing infrastructure. His conclusion is an interpretation of aggregate economic and corporate trends, not proof that AI has delivered no benefit at any individual company or job function.
Apollo’s “Daily Spark” note, titled “No Signs of AI Payoff Outside the Tech Sector,” compares the seven largest technology companies with the other 493 firms in the S&P 500. Apollo’s charts show revenue per worker rising for the seven and remaining broadly flat for the other companies. The technology group’s profit margins also rose to roughly 30%, while the other 493 remained near 10%, according to the note’s reading of the data.
The analysis also cites San Francisco Federal Reserve productivity measures, which it says show no clear AI-driven boost to productivity growth so far. Separate spending figures from Ramp, a corporate card and spending platform with more than 85,000 U.S. business customers, indicate that a small share of firms accounts for most AI-related outlays. The top 10% of firms account for 99.5% of spending on model serving and inference, Apollo reported. Those figures describe Ramp customers, not all U.S. companies.
Meanwhile, Meta’s October 22 regulatory filing lists about $825 billion in future commitments, including leases not yet started and contractual purchases, largely tied to data centers, servers, cloud capacity and network equipment. The commitments run across multiple years and include items not exclusively related to AI. Meta also forecast $135 billion to $145 billion in capital spending for 2026. Separately, the source report says SpaceX borrowed $40 billion to buy AI chips, underscoring the financing scale of infrastructure expansion.
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
AI Spending Meets Uneven Returns
The contrast matters because the costs of building AI capacity are visible in corporate filings, while the returns remain difficult to identify across much of the economy. If companies keep spending on chips, data centers and cloud services before those investments raise revenue or reduce costs, the gap could weigh on margins and increase pressure to show results. Apollo’s Slok has described token spending as a cost center for mainstream companies, rather than a clear source of growth.
That does not mean AI has no economic value. The source report cites studies in which customer-support workers resolved 14% more issues per hour, rising to 34% for less experienced workers. Retail experiments recorded sales changes ranging from zero to a 16.3% increase, depending on the workflow. These are specific settings, not economy-wide estimates, and they show why broad averages can miss practical gains as well as failed deployments.
For workers and consumers, the key question is whether those task-level improvements spread into durable productivity, better services or lower prices. For investors and company leaders, the issue is whether revenue and efficiency gains eventually justify the large infrastructure bills. The evidence cited by Apollo does not yet settle either question.
Popular right now
Vitamix Propel Series 750 Professional-Grade Blender, 4 Programs, Black
Ninja CREAMi 7-in-1 Ice Cream Maker, NC301
Bose QuietComfort Wireless Noise-Cancelling Over-Ear Headphones, Black
As an affiliate, we earn on qualifying purchases.
What Apollo’s “Zero” Measures
Slok’s phrase that AI has delivered “zero financial payoff” is a summary of aggregate trends outside technology, not a direct measurement of every company’s AI projects. The comparison between the seven technology giants and the rest of the S&P 500 shows a difference in revenue per employee and profit margins, but it cannot establish that AI caused that difference. Other factors, including labor costs, interest rates and tariffs, may affect company results.
The note’s evidence combines economic productivity indicators, S&P 500 company data and Ramp spending records. Each has limits: the Ramp sample is not the whole economy, and averages across hundreds of companies can conceal businesses that gained or lost from AI. The source report says nearly 9,000 executives were surveyed, with 83% reporting no productivity impact so far and an average estimated impact of 0.2% a year. Those findings suggest limited reported gains to date, rather than a definitive measure of AI’s long-term effect.
Spending plans also need careful interpretation. Meta’s approximately $825 billion figure is a total of different kinds of future obligations, not a single-year cash bill or an amount entirely devoted to AI. Forecasts that the five largest technology infrastructure spenders could invest nearly $780 billion in 2026, and more than $1 trillion annually from 2027 through 2030, are projections cited in the source report—not commitments already made.
“No Signs of AI Payoff Outside the Tech Sector”
— Torsten Slok, Apollo chief economist, in the October 24 Daily Spark note
The Payoff Is Hard to Isolate
It remains unclear how much of the observed gap between technology companies and other firms is caused by AI. Apollo’s comparisons do not separate AI’s effect from other economic forces, and company-level averages may hide substantial variation. The source material also does not provide enough detail to independently assess the methods behind each workplace experiment or how well its results generalize to other businesses.
The full financial exposure from infrastructure spending is also not equivalent to immediate debt or cash spending. Meta’s commitments include future leases and contracts, while some relate to operations beyond AI. The report says debt insurance costs for Meta and CoreWeave have risen, but does not provide enough detail here to establish the cause or how much financing costs have changed across the sector. The durability of AI demand, the returns on new infrastructure and the pace at which companies will adopt paid services remain unsettled.
Evidence From Earnings and Productivity
The next useful indicators will be company disclosures showing whether AI-related products generate recurring revenue and whether AI tools reduce costs in specific operations. Investors and analysts will also track capital spending, cash flow and financing conditions as major technology firms build out data centers and computing capacity.
For the economy-wide case, future productivity and output data will matter more than investment announcements alone. The present evidence points to a clear divide: large infrastructure outlays are being recorded, while broad returns outside technology have not yet appeared in the measures Apollo cites. Whether that changes will depend on results reported in coming quarters and on whether task-level gains spread across more companies.
Key Questions
What did Apollo conclude about AI’s financial returns?
Apollo chief economist Torsten Slok said the data show no visible broad financial payoff outside the technology sector so far, based on aggregate revenue, margins and productivity indicators.
Does Apollo’s analysis prove AI has produced no benefits?
No. The conclusion concerns broad trends, not every firm or task. The source report cites customer-support and retail studies with gains in some settings, while noting that results vary by workflow.
What does Meta’s roughly $825 billion figure include?
It combines future lease obligations and contractual commitments, mostly related to data centers, cloud capacity, servers and network equipment. It is a multiyear total, not a single-year spending bill, and not all of it is exclusively for AI.
Why are companies spending so much if returns are uncertain?
Companies are investing in computing infrastructure and AI services in anticipation of demand and future uses. The available source material does not establish whether the expected returns will match those costs.
Source: ThorstenMeyerAI.com
