The $250 Billion Mirage

Solow is back. The AI industry should be nervous.

In 1987, Nobel laureate Robert Solow looked at a decade of computer investment and delivered a line that still stings, you can see the computer age everywhere except in the productivity statistics. Forty years later, the same verdict just landed on generative AI.

A February 2026 NBER study of 6,000 executives across the US, UK, Germany, and Australia found nearly 90% of firms report AI has had no measurable impact on employment or productivity over three years. Two-thirds of executives use AI, for about 1.5 hours a week. A quarter don’t use it at all. Meanwhile, corporate AI spending blew past $250 billion in 2024. The money is real. The returns are vapor.

Apollo’s Torsten Slok put it bluntly, AI is everywhere except in the incoming macroeconomic data. No signal in employment. No signal in productivity. No signal in inflation. Outside the Magnificent Seven (Apple, Microsoft, Alphabet, Amazon, Nvidia, Meta, and Tesla), no signal in profit margins or earnings expectations.

The research gets worse. Boston Consulting Group surveyed 1,488 US workers and found a cliff edge, productivity rose with three or fewer AI tools, then collapsed at four or more. They called it AI brain fry. ManpowerGroup tracked 14,000 workers across 19 countries and watched regular AI use climb 13% in 2025 while confidence in the technology dropped 18%. Adoption is rising. Trust is not.

Stanford’s Digital Economy Lab tracked 200,000 households and found generative AI made online chores 76% to 176% more efficient. The time saved was spent on TV and socializing, not on skill development or work output. Efficiency without reinvestment is not productivity. It is leisure.

This is a governance failure at scale. Executives deployed AI without measuring baselines, without defining outcomes, without controls on tool sprawl, without training, and without the workflow redesign that made the 1990s IT surge finally pay off. They bought the pitch deck and skipped the implementation.

Optimists invoke the J-curve, pain now, payoff later. Maybe. The original paradox resolved into real gains by the late 1990s. But that took reorganized work, better data pipelines, and operational discipline. Three things AI rollouts are missing.

Here is the uncomfortable truth for boards and CFOs. If you cannot point to unit economics, measured workflow gains, headcount reallocation, or margin expansion that justifies your AI spend, you are not running a transformation. You are funding someone else’s narrative.

Solow already wrote the epitaph. Leadership should stop ignoring it.

Are we at the point of calling it fraud?? When the bubble bursts are we going to hold the accounting firms accountable as is there any way to not see the outright market manipulation?