Outside of tech there has been no financial payoff to US companies for using AI, now viewed as a margin-hurting cost center, as VC-backed firms burn more tokens

AI investments are no longer automatically generating higher revenues or profit margins for non-tech US companies.

Artificial intelligence has not yet yielded any financial benefits for companies outside the technology sector in the US. While the 'Mag 7' tech companies are seeing increases in revenue and profit per worker, broader S&P 493 companies show flat performance in these metrics. There is also no evidence of AI boosting overall productivity growth across the S&P 493. Consequently, non-tech corporations are reducing AI-related spending, viewing it as a margin-reducing cost rather than a growth driver. Unlike venture-backed startups that are increasing their AI token expenditure, established companies are hesitant to adopt the latest AI models, especially if they don't see a clear return on investment. This limited adoption means AI's macroeconomic impact remains minimal for now.

The current AI landscape shows a significant divergence between leading tech firms and the rest of the corporate world. The 'Mag 7' are leveraging AI for tangible gains in revenue per worker and profit margins, while the majority of S&P 493 companies are experiencing stagnant performance. This suggests that AI’s current benefits are concentrated and not broadly transferable, raising questions about the general applicability of AI in driving core business value.

Mainstream corporations are increasingly scrutinizing AI expenditures, shifting from viewing it as an innovation investment to a pure cost center. This implies that beyond specialized tech applications, AI's adoption hinges on its ability to demonstrate concrete productivity and profit improvements. The critical next step will be observing if and how these companies can translate AI's analytical capabilities into the nuanced judgment and accountability required for sustainable revenue growth and margin expansion, especially as AI models become more commoditized.