Executives are describing AI’s effect on productivity largely as something still to come, according to new research from the Federal Reserve Bank of St. Louis, even as aggregate productivity shows very little actual productivity gains.
What are execs saying about AI?
The St. Louis Fed analyzed about 490,000 earnings call transcripts from 5,198 publicly traded U.S. companies covering 2000 through 2025. Researchers used an open-source large language model to flag every sentence about productivity, determine whether it also mentioned AI and classify whether the speaker was describing a past, present or future gain. The share of productivity-related sentences that also mentioned AI was near zero before ChatGPT’s release, climbed through 2023, plateaued in 2024, then accelerated again in 2025 to about 15% of productivity sentences by year-end.
Across AI-related productivity sentences, about 95% referred to future gains, compared with roughly three-fourths of non-AI productivity sentences. Tone followed the same pattern: 95% of AI-related sentences described productivity as increasing, against 75% for non-AI sentences. Fed researchers Serdar Ozkan, Aakash Kalyani and Nicholas Sullivan write that firms are “investing, experimenting and reorganizing around AI today, while the measurable productivity effects remain mostly ahead.”
Who is most optimistic about AI productivity?
That optimism has not yet shown up in the broader economy. The researchers cite Federal Reserve Bank of San Francisco data showing utilization-adjusted total factor productivity grew just 0.07% over the four quarters ending in the first quarter of 2026. A related San Francisco Fed study by Kalyani and Huiyu Li found that firms with positive AI sentiment on earnings calls have posted higher capital spending and research and development growth than other public companies. They note this combo of high positive sentiment and high spending is concentrated among large technology firms building AI infrastructure.
The researchers say they will continue tracking whether the language in earnings calls shifts from expectations to realized results in the coming quarters. For HR leaders, the research offers a useful counterweight to internal narratives that treat AI productivity gains as realized. Workforce plans, hiring freezes or layoffs built on the assumption that AI has already delivered efficiency gains are running ahead of what the data show. Executives on earnings calls are themselves describing those gains in the future tense to investors.
Credit: Source link
