Economists Struggle to Gauge AI's 4-Year Economic Impact as Data Sends Conflicting Signals
Updated
Updated · The New York Times · Jul 2
Economists Struggle to Gauge AI's 4-Year Economic Impact as Data Sends Conflicting Signals
3 articles · Updated · The New York Times · Jul 2
Summary
Less than four years after generative AI moved into mainstream corporate use, economists still cannot pin down what it is doing to jobs, inflation or productivity right now.
Conflicting evidence drives that uncertainty: some measures link AI to high unemployment among new graduates and tens of thousands of lost jobs, while others suggest companies are hiring more because of it.
Government statistics are a poor fit for the shift because they are backward-looking and better at capturing broad trends than fast changes in specific sectors, regions or newly emerging industries.
Researchers still disagree on basic facts such as how many companies use AI and which workers are most exposed, raising the risk that policymakers will get clear data only after the disruption is well underway.
With AI's economic impact a mystery, how can leaders decide anything when the data will only arrive too late?
AI skills now command a huge wage premium, but for how long before AI learns those same skills?
Is preparing society for AI harm a better strategy than trying to control the technology's development itself?
AI’s $1 Quadrillion Question: Conflicting Forecasts and the Four-Year Economic Impact (2026 Report)
Overview
As of July 2026, economists are struggling to measure the true economic impact of Artificial Intelligence (AI) over the past four years. There is major uncertainty about how AI will shape the global economy, with forecasts for annual growth ranging from just 0.1% to as high as 30%. This huge gap could mean a quadrillion-dollar difference in US GDP by 2035. The main reason for this uncertainty is a sharp disagreement between traditional economists, who expect modest growth from AI, and optimistic AI practitioners, who predict much higher gains. This divide makes it difficult for policymakers and businesses to plan for the future.