Date
Publisher
arXiv
Does generative artificial intelligence (AI) widen or narrow productivity gaps across workers? We study this in a randomized online experiment with 1,174 adults aged 25-45 who completed a workplace-style problem-solving task with or without a generative AI assistant, followed by an unassisted module. AI improves performance for all participants, but gains are larger among those with less education. Without AI, higher-education participants outperform lower-education participants by 0.548 standard deviations; with AI, the gap falls to 0.139, closing about three-quarters of the initial difference. Chat logs show that lower-education participants obtain substantial assistance, while higher-education participants use AI more effectively. Gains are not purely due to delegation: treated participants do not perform worse once AI is removed, and lower-education participants retain part of their improvement, although a sizable gap re-emerges. Intensive AI use raises assisted performance regardless of participants' own effort, but follow-up performance improves only when intensive use is combined with sustained effort. Generative AI narrows effective productivity differences in task execution, while human-capital differences continue to shape unassisted performance and tool use.
Study at a glance
- Question: Does generative AI help narrow productivity gaps between workers with different levels of education?
- Setting and data: Online study with 1,174 adults aged 25–45 completing workplace-style problem-solving tasks
- Method: Randomized experiment comparing participants working with or without an AI assistant, followed by an unassisted task
Why it matters
Whether generative artificial intelligence widens or narrows education-based productivity gaps across workers remains an open question.
Key findings
- In an experiment with 1,174 adults, generative AI closed about three-quarters of the performance gap between lower- and higher-educated participants.
- After AI was removed, lower-educated participants retained part of their performance gains, though a sizable gap between education levels re-emerged.
- Heavy AI use boosted assisted performance regardless of effort, but follow-up unassisted performance improved only when paired with sustained personal effort.
Summary drafted with AI from the paper’s abstract and may contain errors. See the full paper for details.
What is the application?
Who is the user?
Who age?
Study design
