A February 2026 issue brief from the International Center for Law and Economics synthesizes the empirical literature on AI productivity and labor markets, finding consistent task-level and firm-level productivity gains of 15% to over 50% across writing, legal, coding, accounting, and customer-support settings, with disproportionate gains for lower-skilled workers producing within-occupation skill compression. Aggregate labor-market indicators through late 2025 show no economy-wide job loss or wage decline, though entry-level workers in highly exposed occupations show localized employment declines of roughly 16% relative to trend. The gap between micro-level gains and flat macroeconomic aggregates is attributed to the productivity J-curve, measurement limitations in GDP statistics, and uneven organizational adoption rather than absent effects. For financial institutions, the review provides a structured framework for assessing where AI-driven cost reduction is most empirically grounded and where macro-level revenue or displacement assumptions remain premature.
397 other tracked developments touching the same claims.