Slok's thesis
Torsten Slok, chief economist at Apollo Global Management, has published a note in which he states that artificial intelligence has not produced any economic benefit for large U.S. companies outside the technology sector. According to TechCrunch, the analysis focuses on a group defined by the economist himself as the “S&P 493”: the S&P 500 excluding Alphabet, Amazon, Apple, Meta, Microsoft, Nvidia and Tesla.
Within this scope, Slok finds profit margins and revenue per worker essentially flat, with no signs of an increase in productivity attributable to AI. His conclusion is that, despite massive investments and media attention, the technology has so far not generated a measurable financial return for corporate America outside tech.
The note does not merely capture a lag: Slok is skeptical of the idea that non-tech companies are simply catching up with the large technology companies. In his judgment, it will be difficult to transform increasingly low-cost intelligence into higher revenue growth and higher margins.
Judgment and reporting
The central point of the argument is qualitative in nature. Slok writes that AI can be an excellent calculator, but in business intelligence is above all judgment. He adds that the technology does not encompass responsibility, trust, culture, unwritten rules, ethical trade-offs or fairness—factors that in his view affect reputational risk and, ultimately, drive growth and margins.
TechCrunch links this assessment to other recent studies according to which AI has not yet produced significant productivity gains. This is a convergence of indications, not definitive proof: the available material does not detail the methodology of Apollo's note or the data underlying the calculation of margins and revenue per worker.
What remains to be verified
The news has signaling value because it comes from an economist at a major financial firm and concerns a broad universe of U.S. public companies. However, several questions remain open that the material does not clarify: the time horizon of the analysis, the comparison with previous periods, and how the AI investments of the companies considered are measured.
A further distinction must be drawn between what is observed data and what is interpretation: flat margins are presented as empirical evidence, while the considerations on judgment, trust and culture are the economist's assessments. The seven companies excluded from the calculation are indicated as a comparison group, not as the object of analysis. Without detailed data, the thesis of a temporary lag by non-tech companies relative to Big Tech cannot be either confirmed or ruled out on the basis of this material alone.
