AIThis post was created with the assistance of artificial intelligence (AI).

🔍 Read the full analysis: AI Spending Is Growing Fast. What Are Businesses Getting Back? on ThorstenMeyerAI.com

Buying for a business?Offer from Amazon

Get business pricing on office and shipping supplies

  • Business-only prices and quantity discounts
  • Tax-exempt purchasing
  • Multiple users, one account, clear invoices
As an affiliate, we earn on qualifying purchases.

TL;DR

A report by Apollo chief economist Torsten Slok says recent productivity and financial measures show no broad AI payoff outside major technology companies. Meanwhile, corporate spending and future infrastructure commitments are growing, while specific workplace trials show gains in some tasks. The aggregate data do not prove AI has produced no returns for every company.

Corporate spending on AI infrastructure is rising, but recent aggregate data show no clear productivity or financial payoff outside the largest technology companies, according to an October 24 analysis by Apollo chief economist Torsten Slok. The finding is not proof that AI has generated no value for individual businesses: the data compare broad company groups and do not isolate AI’s effects from other economic factors.

Apollo’s analysis compares the seven largest technology companies—Alphabet, Amazon, Apple, Meta, Microsoft, Nvidia and Tesla—with the other 493 S&P 500 companies. Its charts show inflation-adjusted revenue per worker rising for the seven, while the other companies have remained in a relatively stable range since 2021. Apollo’s figures put the seven near $780,000 per employee in the third quarter of 2026, compared with about $525,000 for the remaining companies. The analysis also shows profit margins near 30% for the seven and around 10% for the rest.

Those comparisons describe company-level outcomes, not a direct test of AI use. Apollo also cited San Francisco Federal Reserve productivity measures that it said show no signs of AI lifting productivity growth. Separately, data from Ramp, a corporate spending platform with more than 85,000 U.S. business customers, indicate that AI-related spending is concentrated: the top 10% of firms account for 99.5% of spending on model serving and inference, and 95.6% of spending on OpenAI and Anthropic, according to the source report.

At the same time, investment commitments are substantial. Meta’s third-quarter release reported $34.91 billion in capital expenditures and $32.79 billion in operating cash flow, resulting in negative free cash flow of $2.12 billion for the quarter. In its filing, Meta listed about $825 billion in future commitments, including leases and contractual purchases largely tied to data centers, servers and cloud capacity. The figure is not all AI spending, and it covers obligations extending over several years. Meta forecast $135 billion to $145 billion in capital spending for 2026.

At a glance
reportWhen: Apollo analysis published October 24, 2…
The developmentApollo’s October 24 analysis of productivity and corporate results renewed questions about whether fast-growing AI investment is delivering measurable returns across businesses.
Thorsten Meyer AI
Apollo · 20–24 Oct 2026

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.

≈$780kRevenue per employee, Magnificent 7, Q3 2026 (inflation-adjusted, read off Apollo’s chart)
≈$525kRevenue per employee, S&P 493: inside the band it has held since 2021
≈30%Profit margin, Magnificent 7, Q3 2026 (S&P 493: ≈10%, flat)
80.4%of the top 1% of AI spenders are venture-backed (median spender: 13.1%)

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

  • Model serving & inferenceAI99.5%
  • NeocloudAI99.0%
  • Cloud computing95.7%
  • OpenAI + AnthropicAI95.6%
  • Paid ads91.9%
  • SaaS excluding AI91.8%
  • CRM84.2%

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

The Cost of AI Without Broad Returns

The gap between large infrastructure bills and uncertain company-wide returns matters to executives, investors and workers. Businesses must decide whether to expand AI use, how to pay for computing capacity and how to judge whether deployments are improving output enough to justify their costs. High spending by a small group of companies can also make industry-wide investment figures look more representative of business adoption than they are.

The aggregate picture is not the whole story. The source report cites research showing customer-support workers resolved 14% more issues per hour with AI assistance, with a 34% gain among novice workers. Retail experiments reportedly ranged from no sales increase to a 16.3% rise, depending on the workflow. Those results suggest benefits can appear in specific tasks, but they do not establish that the same gains will carry across industries or offset the cost of AI systems.

Slok’s analysis also cites a survey in which 83% of nearly 9,000 executives said AI had not yet affected productivity at their organizations; the reported average productivity impact was 0.2% a year. The survey result is a snapshot of respondents’ assessments, not a universal measurement of every company’s performance.

Products Worth Considering

Canon SELPHY CP1500 Wireless Photo Printer Bundle with 108 Sheets of KP-108 Paper, 3 Ink Cartridges, and Cleaning Cloth

Canon SELPHY CP1500 Wireless Photo Printer Bundle with 108 Sheets of KP-108 Paper, 3 Ink Cartridges, and Cleaning Cloth

Apple MacBook Air 15-inch Laptop with M5 Chip, 16GB Memory, 512GB SSD, Liquid Retina Display, and Touch ID (Midnight)

Apple MacBook Air 15-inch Laptop with M5 Chip, 16GB Memory, 512GB SSD, Liquid Retina Display, and Touch ID (Midnight)

As an affiliate, we earn on qualifying purchases.

What Apollo’s Comparison Measures

Apollo’s October 24 note, titled “No Signs of AI Payoff Outside the Tech Sector,” brought together company financial measures, productivity data and spending information. Its central comparison is between the Magnificent Seven and the rest of the S&P 500. That framing can show that results are concentrated, but it cannot determine how much of the difference was caused by AI rather than existing business models, investment cycles, labor costs, interest rates or other forces.

The spending data have a different scope. Ramp’s figures represent its own customer base, while Meta’s filing records its company-specific obligations. Apollo’s analysis says that 80.4% of the top 1% of AI spenders are venture-backed, compared with 13.1% of median spenders. That suggests heavy AI expenditure is concentrated among firms with access to venture funding; it does not show that those expenditures have produced financial returns.

The infrastructure buildout also extends beyond Meta. The source report cites projections that Amazon, Alphabet, Meta, Microsoft and Oracle together could spend nearly $780 billion in 2026, with annual spending above $1 trillion from 2027 through 2030. These are forecasts, not reported commitments or completed expenditures. Meta’s $825 billion figure, by contrast, is a total of future obligations disclosed in its filing, not a single-year budget.

Where the Return Evidence Falls Short

How much AI itself contributed to the reported revenue, margin and productivity trends remains unclear. Company averages can conceal businesses that gained or lost, while the cited measures do not separate AI’s impact from other changes in the economy. Ramp’s spending figures also cover Ramp customers, not the S&P 500 as a whole.

It is also not clear whether early task-level improvements will persist at scale or exceed the full cost of software, computing, implementation and staff training. The source report gives examples of measured gains in customer support and retail, but does not provide enough detail here to compare their methods, duration or costs. Meta’s future commitments include non-AI items and span multiple years, so they should not be read as an annual AI bill.

The Next Test Is Measurable Output

Investors and business leaders will be watching upcoming company filings and productivity data for evidence that AI spending is translating into higher output, revenue or margins. For individual companies, the clearest test is likely to be whether a defined AI-supported process produces repeatable improvements that exceed its total operating costs.

Meta’s next financial disclosures should show how its capital spending and cash flow develop against its 2026 outlook. Broader evidence will take longer: productivity measures and company results need to show sustained changes before analysts can distinguish a durable AI effect from short-term investment and economic conditions. The available data establish that investment is growing and returns remain uneven; they do not settle how much value AI will generate over time.

Key Questions

Did Apollo prove that AI has produced no financial returns?

No. Apollo’s conclusion reflects aggregate indicators that show no clear broad payoff outside major technology companies. The analysis does not directly measure AI’s effect at each company or rule out gains in particular tasks.

What does Meta’s $825 billion figure represent?

It is an estimate of future commitments disclosed in Meta’s filing, including leases and contractual obligations largely related to data centers, servers and cloud capacity. It spans multiple years, includes some non-AI items and is not a one-year spending total.

Are there examples of AI improving business performance?

The source report cites a customer-support study showing 14% more issues resolved per hour, including a 34% gain for novice workers, and retail experiments with sales results ranging from no increase to 16.3%. These are specific findings, not guarantees of similar results elsewhere.

Why can’t the aggregate figures isolate AI’s effect?

Revenue, margins and productivity can change for many reasons, including labor costs, interest rates and business mix. A comparison between large technology companies and other firms cannot by itself show which changes were caused by AI.

Source: ThorstenMeyerAI.com

This content is for general information only and is not financial, tax or legal advice. Consult a qualified professional for decisions about your money.
You May Also Like

The CFO’s new operating system. Anthropic, OpenAI, and the consulting margin that just got compressed.

Anthropic’s $1.5B joint venture and OpenAI’s parallel funding mark a shift to integrated AI operating systems for enterprise finance, replacing traditional consulting models.

Slow To Embrace AI, Difficult To Dislodge Later

Analysis of how slow AI adoption by enterprises creates durable incumbents that are hard to dislodge, challenging common assumptions about disruption.

FCC says it will move toward 2027 auction of mid-band wireless spectrum

FCC announces plans to move toward a 2027 auction of mid-band wireless spectrum, impacting 5G deployment and telecom industry strategies.

Understanding Anthropic’s $965B Series H: The Compute Revolution

Anthropic’s $965 billion valuation is a strategic investment in AI infrastructure, focusing on chips, memory, and power to enable next-generation models like Claude.