Artificial intelligence has delivered "zero financial payoff" to corporate America outside technology, according to Torsten Slok, chief economist at Apollo Global Management. His review of the S&P 493 shows profit margins and revenue per worker remaining flat despite massive investment and widespread hype. For business leaders, the message is that spending on AI does not yet translate into measurable financial returns.
What Slok found in the S&P 493
Slok defines the S&P 493 as the S&P 500 excluding Alphabet, Amazon, Apple, Meta, Microsoft, Nvidia and Tesla. That separation isolates large non-tech corporations from the seven technology producers and heavy AI investors. Within that group, he reports no visible improvement in productivity tied to AI adoption. Margins have not expanded and output per employee has not moved upward.
The mechanics of the argument rest on two financial indicators: profit margin and revenue per worker. Both are treated as proxies for whether cheaper intelligence turns into growth and efficiency. If AI tools were reshaping operations, at least one of these metrics should have shifted across hundreds of companies. Slok finds neither has, which leads him to describe the payoff as zero rather than delayed or uneven.
The note lands against a background where other recent studies also report no significant productivity gains from AI so far. Non-tech companies have purchased software, run pilots and discussed automation, while vendors have promoted rapid transformation. Slok questions whether laggard adoption explains the gap, expressing doubt that corporate America can easily convert increasingly cheap intelligence into higher revenue growth and higher margins.
What zero payoff means for AI buyers
For companies using or deploying AI, the practical consequence is a stricter test for new projects: licenses, integration and training must be linked to revenue or margin, not to usage. Technology suppliers already capture value from infrastructure and models, while buyers carry implementation costs. Small firms will feel this faster because pilot expenses weigh more heavily, while large firms may absorb longer periods without return across many departments.
The limitation centers on what AI does not handle, in Slok's description: accountability, trust, culture, unwritten rules, ethical trade-offs and fairness. He contrasts calculation with judgment, arguing that reputational risk and commercial decisions depend on factors models do not understand. Buyers should therefore verify where human approval remains required, which processes carry reputational exposure, and what the vendor evidence shows beyond demonstrations.
The marker to watch is movement in revenue per worker and profit margins for the S&P 493 over coming quarters. A sustained rise alongside documented AI deployment would suggest the payoff has started. Continued flat readings would support Slok's position that cheaper intelligence alone does not create growth.
