Data from the Federal Reserve Bank of San Francisco shows that over the past four quarters ending in the second quarter, U.S. total factor productivity (TFP) grew by 1.10%, lower than the 1.61% in the first quarter; utilization-adjusted TFP declined by 0.42%, with the annualized decline reaching 2.19% in the second quarter alone. Aggregate data indeed has not yet shown a productivity leap brought about by AI. The latest research from the Federal Reserve suggests that AI is still in the stages of capability enhancement, capital expenditure, and enterprise adoption; while its application scope is expanding, its intensity is relatively shallow, and efficiency improvements at the task level have not yet been translated into enterprise and macroeconomic output through process reengineering. Historically, there has typically been a time lag of several years between general technology investment and the realization of productivity gains.
