Does AI narrowing the gap between weak and strong workers on one job actually shrink pay inequality economy-wide?

This explores whether AI raising the output of less-skilled workers actually leads to narrower income gaps, or whether the two can come apart. The corpus gives a clear split answer. Inside a single job, AI often does compress the skill gap. Across the whole economy, that compression doesn't automatically reach paychecks, and several models suggest gains may flow to the owners of capital instead.

The within-job evidence is strong and consistent. At a Fortune 500 support center, AI assistance raised productivity 15% on average, and most of that went to newer agents. The most experienced agents saw a slight drop in quality Does AI assistance help less experienced workers most?. Among BCG consultants, bottom-half performers gained the most from GPT-4 Does AI help lower-skilled workers catch up to top performers?. In a randomized experiment, AI closed about three-quarters of the performance gap tied to education, and some of that gain lasted after the AI was taken away Can AI narrow the education performance gap?. There's a catch hidden in the BCG result, though: the leveling held when AI use was mandatory. When people could choose whether to use it, the effect risked reversing. The catch-up depends on who actually adopts the tool. Another note adds a limit: these gains show up when people apply skills they already have, and they disappear when people use AI to learn something new When does AI actually boost worker productivity?.

The surprising part is the distance between "performs better" and "earns more." Acemoglu's task-based model finds AI's wage effects are spread more evenly across demographic groups than earlier automation was. Yet it still predicts capital income pulling ahead of labor income Will AI increase inequality less than past automation?. Anthropic's scenario model goes further. As AI growth speeds up, workers' share of income falls, and knowledge-worker wages can stall even while average wages and GDP rise. Concentrated ownership and the difficulty of moving between occupations block broad sharing Does AI growth inevitably shift wealth away from workers?. In other words, a low-skill worker can become nearly as good as a senior colleague while the money from that improvement goes to the firm. Some workers seem to sense this. In a survey of 81,000 Claude users, the people with the biggest AI speedups were among the most worried about losing their jobs Does AI productivity gain always ease job displacement fears?. Many workers also hide their AI use from colleagues and managers Why do workers hide productivity gains from AI use?, so even real gains may never show up in pay talks.

There's also a long-run twist that few people expect. An overlapping-generations model shows that automating entry-level work can raise output today while slowing growth. It pulls novices away from the experts who pass on tacit know-how Can automation raise output while slowing growth?. If AI is what lifts the low-skill worker, the usual path from junior to senior may erode. That path is how low-skill workers used to become high earners. The help that narrows today's gap could shrink tomorrow's ladder.

The corpus's bottom line is that the technology doesn't settle the outcome. An interdisciplinary review across work, education, health, and information concludes that generative AI can push inequality in either direction. Access, integration, and incentives decide which Does generative AI inevitably worsen or reduce inequality?. Measurement also blurs things. Overall productivity gains so far look small, around 1% What is generative AI's actual aggregate productivity gain today?, and early AI investment may hide real gains before they show up Why does AI productivity appear weak when investment is high?. So yes, productivity gains for low-skill workers can reduce inequality, but only if policy, bargaining power, or ownership structures carry those gains from the task into the paycheck.