Governor Barr warns AI could widen or narrow U.S. inequality
The speech highlights AI's potential impact on U.S. workers and households, urging policymakers to consider equity as adoption expands.
Governor Barr said AI could reshape income and wealth distribution, urging policymakers to address potential inequality as the technology spreads.
He noted that in 2024 the highest-earning one-fifth of households captured 52 percent of all income while the bottom 20 percent earned only 3 percent, and that the United States ranked sixth most unequal among G20 members. Wealth is even more concentrated: the bottom one-half hold less than 3 percent, the top one-tenth hold 59 percent, and the top one-tenth of a percent hold 15 percent.
Barr outlined scenarios in which AI could widen gaps. A Federal Reserve Survey found 43 percent of workers with graduate degrees used AI in the prior month versus 10 percent of those with a high-school education or less, suggesting higher-paid workers may reap more benefits. He cited research indicating AI may displace labor, especially for new entrants, and warned that economies of scale and data advantages could concentrate market power among a few "hyperscalers," as described in a 2025 paper.
Conversely, he described how AI might reduce inequality by democratizing capability. An experiment showed AI cut average task time by 40 percent and raised output quality by 18 percent, with the greatest gains for lower-performing participants, suggesting productivity gains could be broadly shared.
Barr concluded that while the Federal Reserve does not set policy on AI, other policymakers should consider education, job training, competition, tax policy and related measures to ensure AI's benefits are widely distributed rather than confined to a small group of firms and investors.