Macro Signal Detected
Stanford HAI's 2026 AI Index, the most comprehensive annual survey of AI's societal and economic impact, reports for the first time that AI adoption is producing statistically measurable labor productivity gains at the macroeconomic level in four countries: the United States, the United Kingdom, South Korea, and Sweden. The researchers found productivity growth approximately 0.4-0.6 percentage points above trend in knowledge-worker-heavy sectors in these countries, attributable to AI tool adoption.
The Methodology
Isolating AI's contribution to productivity growth requires controlling for other factors (capital investment, education, economic cycle). The Stanford team used a difference-in-differences approach comparing productivity growth in industries with high AI tool adoption versus low-adoption industries within the same countries, controlling for pre-existing trend differences.
Where Gains Are Concentrated
Software development shows the largest measured productivity gain — approximately 25-35% in lines of code produced per developer-hour for teams that have adopted AI coding assistants at 60%+ usage rates. Legal services show 15-20% productivity gains on document-intensive tasks. Customer support shows 30-40% throughput improvement. Creative services (writing, design) show more modest 10-15% gains, with quality effects harder to measure.
The Transition Period Risk
The report also documents a polarization effect: workers who adopt AI tools see productivity gains, while workers in the same field who do not adopt experience relative productivity declines that can translate to employment risk. The transition period — as slower adopters catch up or exit — may be accompanied by meaningful labor market disruption in affected sectors.