In 1987, Robert Solow looked at the computer revolution and observed: “You can see the computer age everywhere but in the productivity statistics.” Nearly 40 years later, Apollo chief economist Torsten Slok is making the same observation about AI. Sasha Rogelberg reports in Fortune on new data that makes the parallel hard to ignore.

Among 6,000 executives surveyed, 90% said AI has had no impact on employment or productivity over the last three years. Average executive AI usage: 1.5 hours a week. That’s barely trying.

Slok, echoing Solow:

AI is everywhere except in the incoming macroeconomic data. Today, you don’t see AI in the employment data, productivity data, or inflation data.

The Solow paradox eventually resolved itself. Computers didn’t show up in productivity stats until the mid-1990s—decades after they entered the workplace. The technology arrived long before organizations figured out how to restructure around it.

Slok sees the same pattern forming:

The value creation is not the product, but how generative AI is used and implemented in different sectors in the economy.

That’s the part most companies are skipping. They’re giving employees an AI chatbot and expecting the productivity graph to move. The companies where AI is actually changing output are the ones rethinking their workflows. Most stall at the tooling.

If the Solow parallel holds, the productivity gains are coming. They’ll show up first in the companies that did the reorganization. I have a feeling that this Claude Code trend is going to hold and show up in stats next year.