Recommended Articles
Markets & Policy

Apollo's Torsten Slok: AI Has Delivered 'Zero Payoff' Outside the Magnificent 7

By Angeled · October 26, 2026 · Markets & Policy
Apollo's Torsten Slok: AI Has Delivered 'Zero Payoff' Outside the Magnificent 7
Apollo's Torsten Slok: AI Has Delivered 'Zero Payoff' Outside the Magnificent 7

In a note published Saturday, October 24, 2026, in Apollo's Daily Spark, chief economist Torsten Slok made one of the sharpest claims of the AI era: "Despite massive investments and widespread hype around artificial intelligence, the technology has delivered zero financial payoff to corporate America outside of tech." The claim, reported by TechCrunch on October 25, is an economist's argument built on index-level data — not a settled fact — but it is the kind of argument that moves boardroom conversations, and founders should understand exactly what it says and what it does not.

The S&P 493: how Slok slices the market

Slok's device is simple. Take the S&P 500 and remove seven companies — Alphabet, Amazon, Apple, Meta, Microsoft, Nvidia, and Tesla, the so-called Magnificent 7. What remains is the "S&P 493." Then compare the two groups on the metrics that should show an AI payoff: revenue per worker, profit margins, and productivity growth.

On Slok's charts, the Magnificent 7 look like a different economy. Inflation-adjusted revenue per employee for the seven giants has risen to about $780,000, while the S&P 493 sits around $525,000 and has not moved. Profit margins tell the same story: close to 30% for the seven giants, about 10% for the other 493 firms, roughly where they were in 2019. And on total factor productivity — measured with the San Francisco Fed's utilization-adjusted series — there is no acceleration despite the AI spending boom.

Slok also documents concentration. The top 10% of firms account for 99.5% of model serving and inference spending, 99% of neocloud spending, over 95% of spending tied to OpenAI and Anthropic, and 95.7% of cloud computing. Venture capital dominates the biggest spenders: 80.4% of the top 1% of AI spenders are VC-backed, versus 13.1% of the median spender.

Why mainstream companies are pulling back

Slok's second claim is behavioral. "Mainstream corporate America is pulling back and starting to view token spending not as a source of growth but as a cost center that weighs on margins," he wrote. In his framing, most companies do not need the latest model and will not pay frontier prices for it. "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 added.

His explanation for the gap is philosophical as much as economic: "AI may be a great calculator, but in business, intelligence is mostly judgment. 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." In other words, the hard part of business is not computation — it is the judgment-laden work that AI cannot yet own.

What the argument gets right, and where to be careful

Slok's framing is powerful because it is falsifiable and specific: if AI were transforming the broad economy, margins and revenue per worker outside tech should be rising. They are not, on his data. That is a genuine challenge to the "AI will lift all boats" narrative, and it echoes other recent studies finding that AI has yet to deliver significant productivity gains across the economy.

But three caveats matter. First, the Magnificent 7 are not "tech" in a narrow sense — Amazon and Tesla are retailers, logistics operators, and manufacturers as much as technology companies, so the split is cruder than the label suggests. Second, index-level margins are noisy: they reflect energy prices, interest rates, tariffs, and sector mix, not just AI. Third, technology adoption historically shows a lag — electricity and the internet both took years to show up in productivity statistics, a pattern economists call the productivity paradox. Flat numbers today do not prove flat numbers forever.

What it means for founders

The practical lesson is not "AI is useless." It is that buying AI is not the same as benefiting from it. Slok's data suggest the payoff so far has gone to companies selling AI, not companies buying it. For founders, that reframes the pitch: do not sell "AI adoption" as an outcome. Sell a specific workflow changed, a cost removed, or a revenue line created — and be ready to show it in the customer's own margins, not in demo metrics.

It also sharpens the fundraising question. If mainstream companies treat token spending as a cost center, startups selling AI tools need to prove they reduce total cost or increase revenue per worker quickly, or they will be cut in the next budget cycle. The winners in the next phase will be the companies that turn cheap intelligence into measurable business judgment — not the ones that simply resell model access.

Angeled.com connects founders with investors.

Source: techcrunch.com
TechCrunch, "AI delivers 'zero financial payoff' to most companies, Apollo economist says" (October 25, 2026)
Source: tradingkey.com
TradingKey/Cryptopolitan, "Apollo's Torsten Slok sees zero AI payoff outside the Magnificent 7" (October 26, 2026)
Source: linkedin.com
LinkedIn News, "Apollo's Chief Economist: AI has delivered 'zero payoff'" (October 25, 2026)

Read Next