AI's $250 Billion Gamble: Productivity Still Lags

Artificial intelligence has attracted heavy corporate spending and frequent mentions in boardrooms, but broad-based productivity improvements have yet to show up across most industries, according to recent research and corporate disclosures. Economists have compared the gap between enthusiasm and measurable results to the “productivity paradox” that accompanied the Information Technology (IT) revolution in the 1980s.

A February study by the National Bureau of Economic Research surveyed 6,000 executives in the U.S., U.K., Germany, and Australia. It found that nearly 90% of firms said AI had produced no impact on employment or productivity over the past three years. The findings sit alongside a sharp rise in corporate outlays, with investment in AI exceeding $250 billion in 2024.

The pattern echoes a well-known historical reference point from the late 1980s. In 1987, Nobel laureate Robert Solow remarked that the IT era’s advances were not visible in productivity statistics. During that earlier period, productivity growth slowed from 2.9% between 1948 and 1973 to 1.1% after 1973, underscoring how transformative technologies can take time to register in economy-wide measures.

Corporate communications suggest AI remains a prominent strategic theme even as aggregate productivity effects remain limited. An analysis found that 374 S&P 500 companies mentioned AI positively in earnings calls between September 2024 and 2025. However, the source material indicates these adoptions have not yet translated into broader economic productivity improvements.

Executives surveyed by the National Bureau of Economic Research nevertheless expressed confidence about future gains. The study reported that firms expect AI to lift productivity by 1.4% and increase output by 0.8% over the next three years. Those expectations were recorded despite evidence of constrained day-to-day use: two-thirds of executives said they use AI for only about 1.5 hours per week, while 25% reported no AI usage at all.

For global markets and policymakers, the disconnect between investment levels and measurable outcomes matters because productivity growth is closely watched as a driver of corporate earnings potential and longer-term economic performance. The study’s cross-country executive sample spans major advanced economies, and the investment figure highlights the scale of capital being directed toward AI.

At the same time, economists are questioning when—or whether—these expenditures will generate tangible returns beyond the technology sector, leaving uncertainty around the timing and breadth of any economy-wide payoff.