Apollo's Slok: No AI Payoff Outside Big Tech

Apollo Global Management chief economist Torsten Slok said on Oct. 24 that artificial intelligence has so far "delivered zero financial payoff to corporate America outside of tech." In a note in Apollo's Daily Spark series, he argued that revenue per worker and profit margins are rising at the seven largest tech companies but are flat across the rest of the S&P 500.

Key takeaways

• Slok says revenue per worker and margins are climbing for the "Mag 7" but are flat for the other 493 S&P 500 companies.
• He sees no sign yet of AI lifting U.S. productivity growth and says mainstream firms are starting to treat token spending as a cost.
• Separate data show adoption is broad but spending is thin: Ramp says 55.9% of U.S. businesses paid for AI in September, at a median of $15.70 per employee a month.

What Apollo's Slok says about AI payoff

The note, titled "No Signs of AI Payoff Outside the Tech Sector," rests on five charts. Two compare the Magnificent Seven with the "S&P 493," the rest of the index. Revenue per worker, adjusted for inflation to the fourth quarter of 2019, moves up for tech and not for everyone else. Profit margins show the same split, based on Bloomberg and Macrobond data.

A third chart uses the San Francisco Fed's utilization-adjusted total factor productivity series and finds "no signs of AI boosting productivity growth." The note does not give the underlying values in its text.

Slok wrote that mainstream companies are beginning to "view token spending not as a source of growth but as a cost center that weighs on margins." Venture-backed firms are spending more on tokens, he said, but most companies do not need the latest models and are unwilling to pay for them.

Is AI spending concentrated in a few firms?

Slok's last two charts draw on Ramp, the corporate card and spend-management company. One shows AI spending is "extremely concentrated in a small number of firms." The other is titled "Circularity in AI shows up in token spending too."

Ramp's own public figures show how wide but shallow adoption is. Its AI Index, built from card, invoice and ACH transactions at more than 85,000 businesses, says 55.9% of U.S. businesses paid for an AI product in September 2026. That is up from 55.2% in August and 12.1 points higher than a year earlier.

Ramp's separate token-spend sample, which the company says is not representative of all its AI spending, showed total token spending up 23.0% year over year in the week ending Oct. 11, while token volume rose 52.4%. Spending is growing, but more slowly than usage.

Context: the productivity data so far

Official statistics have not settled the question. The Bureau of Labor Statistics said on Sept. 3 that nonfarm business productivity rose at a 1.4% annual rate in the second quarter of 2026 and 2.2% from a year earlier, as output grew 2.5% and hours worked rose 0.2% over the year. Those figures do not isolate AI's contribution.

Slok's argument is also partly qualitative. He wrote that "AI does not understand accountability, trust, culture, unwritten rules, ethical trade-offs or fairness," the factors that shape reputational risk and revenue at most companies.

What's next

Slok wrote that while agentic AI and token spending remain confined to a small part of the economy, "the macroeconomic effects of AI will be very limited." His conclusion: "The bottom line is that the AI story will be decided in the S&P 493, not the Mag 7." Third-quarter earnings reports and the BLS's next productivity release will give investors fresh data to test that view.

FAQ

What is the "S&P 493"?
It is shorthand for the S&P 500 minus the Magnificent Seven, the largest tech companies that dominate AI investment.

Does Apollo say AI has no value?
No. Slok says the gains so far show up in tech companies' revenue per worker and margins, not yet in the broader corporate sector or in productivity data.

How many businesses pay for AI?
According to Ramp's AI Index, 55.9% of U.S. businesses in its data paid for AI tools in September 2026. Ramp says its figures likely underestimate actual adoption.

References

• No Signs of AI Payoff Outside the Tech Sector — Apollo, The Daily Spark (Oct. 24, 2026)
• Ramp AI Index — Ramp Economics Lab (September 2026 data)
• Productivity and Costs, Second Quarter 2026, Revised — U.S. Bureau of Labor Statistics (Sept. 3, 2026)