Apollo's Chief Economist Counts the AI Payoff Outside Tech: Zero, With Token Spending Now Booked as a Cost Center

One of Wall Street's most-watched AI skeptics has put a number on the disappointment. Torsten Slok, partner and chief economist at Apollo Global Management, published a note on October 24 arguing that, measured by the only yardstick that matters to most American companies, artificial intelligence has so far produced nothing. "Despite massive investments and widespread hype around artificial intelligence, the technology has delivered zero financial payoff to corporate America outside of tech," Slok wrote in the firm's Daily Spark series.

The evidence is three charts built from Bloomberg, Macrobond and Federal Reserve data. Revenue per worker is rising for the Magnificent Seven, the group of Alphabet, Amazon, Apple, Meta, Microsoft, Nvidia and Tesla, but remains flat for the other 493 companies in the S&P 500. Profit margins tell the same story, expanding for the tech giants and flat for everyone else. And productivity shows no inflection at all: Slok cites the San Francisco Fed's utilization-adjusted total factor productivity series, which has yet to register the acceleration a genuinely general-purpose technology should produce.

The most pointed section of the note concerns where the money is actually going. Drawing on the Ramp AI Index, Slok argues that mainstream corporate America has started treating token spending not as a source of growth but as a cost center that weighs on margins, while venture-backed startups keep increasing consumption because they are building products on the newest models. Roughly 80 percent of the top one percent of AI spenders are VC-backed, he notes, which means a large share of today's compute demand is funded by fundraising rounds rather than operating profits. "As long as agentic AI and token spending are confined to a small part of the economy, the macroeconomic effects of AI will be very limited," he writes.

Slok's deeper argument is that cheaper intelligence does not automatically become fatter margins. "AI may be a great calculator, but in business, intelligence is mostly judgment," he writes, adding that AI lacks the social faculties commercial trust runs on. "AI does not understand accountability, trust, culture, unwritten rules, ethical trade-offs or fairness, all of which shape reputational risk and ultimately drive revenue growth and profit margins." His bottom line: "The AI story will be decided in the S&P 493, not the Mag 7, and so far, there is no evidence that revenue growth or margins are improving outside the tech sector."

The note lands on a nervous market. Meta's five-year credit default swaps widened 7.6 basis points to a record 114.79 basis points on Thursday, CoreWeave's CDS spiked to around 1,145 basis points, and hyperscaler capital spending is running near $800 billion this year with projections above $1 trillion next year, so the gap between spending and measurable returns is exactly where credit investors are looking for cracks.

The caveats deserve equal billing. Productivity statistics have historically lagged technology diffusion by years, electrification and computing followed the same pattern, and flat margins can reflect competition rather than absent gains. Slok is also a credit investor at a firm with a documented bubble-skeptic streak, and his prior warnings have colored his framing. But the discipline of his challenge is hard to dismiss: if AI is a general-purpose technology, the proof will eventually show up in the income statements of hospitals, manufacturers, retailers and banks, not in the earnings of the companies selling the shovels. So far, it has not.