Who Controls the Gains from Workplace AI? Time, Targets and Pay

Abstract

Workplace AI changes the use of paid time before it necessarily changes workers' pay. This commentary synthesizes recent experiments and worker surveys alongside earlier implementation cases to explain how efficiency gains reach employees. Three allocation patterns emerge. In a Danish study, 85% of chatbot users reported redirecting saved time to other job tasks. In a randomized experiment across 66 firms, access to an assistant reduced weekly email time by approximately 1.4 hours and application activity outside usual hours by about a quarter of an hour. Earlier organizational cases document pay increases where more demanding work activated a collective agreement's wage provisions, as well as unchanged pay where job classifications remained stable. These results identify task reassignment, the working-day boundary and wage-setting rules as distinct channels through which gains are distributed. The evidence supports a practical judgment: employee participation in AI adoption should extend to target revision and the treatment of new responsibilities in compensation. Agreements specifying these decisions give workers an explicit claim on the benefits of implementation. The synthesis explains why productivity improvements, time benefits and stable average wages can coexist, and why the allocation rules attached to deployment deserve a central place in workplace AI policy.