22 OCT 2026 (THU) 15:00-16:30

2 days ago

1 min read

Departmental Research Seminars Series

Beyond Exposure: Early Evidence on AI, Creative Quality and Employment

Date: 22 OCT 2026 (Thursday)

Time: 15:00-16:30 (HKT)

Venue: Social Sciences Chamber, Faculty of Social Sciences, 11/F, The Jockey Club Tower, Centennial Campus, HKU

Registration link: https://hkuems1.hku.hk/hkuems/ec_hdetail.aspx?guest=Y&ueid=109848

Abstract:

Nearly four years have passed since breakthrough large language models became publicly available in late 2022. Early assessments inferred labor-market effects from tasks AI could technically perform; enough time has now elapsed to examine how workers and firms adopt these tools and how their responses shape productivity, quality and employment.

Jie Gong (HKU Business School) reports an experiment with professional and student illustrators. AI assistance accelerates initial improvement but reduces there turns to sustained effort, flattening the relationship between time and output quality. Many participants consequently save time rather than pursue the highest attainable quality. Generative AI may therefore improve efficiency while weakening incentives for creative excellence.

Patrick Adler (Department of Geography, The University of Hong Kong) examines whether occupational AI exposure predicts creative employment change between2022 and 2025. Highly exposed creative occupations grew nationally, while similarly exposed metropolitan economies followed divergent trajectories.Interviews with Hollywood post-production workers show that AI replaces bounded tasks but creates complementary work in supervision, correction and quality control. Outcomes also depend on industry organization, production bottlenecks and workplace incentives.

Together, the studies move beyond exposure to show how AI’s consequences emerge through workers’ choices, firms’ organization of tasks and regional production systems.