Apollo's Chief Economist: AI Has Delivered Zero Financial Payoff to the S&P 493
Torsten Slok, chief economist at Apollo Global Management, published an insight note this week arguing that AI has produced no measurable financial return for corporate America outside the technology sector. The analysis focuses on what Slok calls the S&P 493 — the S&P 500 stripped of Alphabet, Amazon, Apple, Meta, Microsoft, Nvidia, and Tesla — and finds both profit margins and revenue per worker are flat, with no sign of AI-driven productivity improvement.
“Despite massive investments and widespread hype around artificial intelligence, the technology has delivered zero financial payoff to corporate America outside of tech,” Slok wrote.
The note is notable for its source. Apollo Global Management oversees more than $1 trillion in assets and advises institutional investors on capital allocation. Slok’s audience is not skeptics looking for ammunition against AI — it is large allocators deciding how to weigh AI-exposed companies in portfolios. A chief economist at a firm of that scale publishing a blunt “zero payoff” conclusion carries different weight than a think-tank paper.
The Argument
Slok is skeptical that the gap will close easily. He argues the structural problem is not lag — that non-tech companies simply need more time — but that AI’s particular strengths don’t map cleanly to the mechanisms that drive profit in most industries.
“AI may be a great calculator, but in business, intelligence is mostly judgment,” he wrote. “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.”
The distinction matters. Most productivity projections model AI as a cost-reduction tool, trimming labor or speeding throughput. Slok’s framing is that revenue growth and margin expansion in most industries depend on judgment-laden relationships — with customers, regulators, suppliers, employees — where current AI provides less leverage than advertised.
Context
The S&P 493 framing is itself a pointed choice. The seven excluded companies have both driven and directly benefited from AI investment: they build the models, sell the compute, and integrate AI into products at scale. Excluding them removes the entities for whom AI has delivered financial returns by construction, leaving a cleaner test of whether AI is generating value for the businesses that buy and deploy it.
Multiple research efforts over the past year have reached similar conclusions on productivity. An NBER working paper published in February found that nine in ten executives across nearly 6,000 companies reported no measurable AI productivity impact after three years of adoption. A Sequoia analysis in July noted that the gap between AI infrastructure investment and AI-derived revenue had widened to an estimated $3 trillion cumulatively.
Slok’s note adds a specific data point from a major institutional perspective: the financial metrics that large investors actually watch — margins, revenue per worker — have not moved for the S&P 493 despite years of AI adoption and billions in enterprise software spending.
The counterargument is adoption lag: AI investment of the past 18 months hasn’t had time to compound into measurable financial outcomes. That argument becomes harder to sustain as the timeframe extends and the models keep getting cheaper and more capable without the productivity numbers arriving.