Generative AI, health risk perception bias and sports participation in older Chinese adults
In two survey waves from 2020–2023, generative AI was linked to more sports participation among older adults, partly via reduced health risk perception bias; this is an abstract only. [1]
Automatic summary · not editorially reviewed
The key findings
The study used two waves of panel data from the China Longitudinal Aging Social Survey spanning 2020–2023. [1]
Generative AI was reported to significantly promote sports participation among older adults, holding after robustness checks. [1]
The corrective effect on health risk perception bias was stronger for self-comparison than for peer comparison. [1]
The source at a glance
- Study type / format
- Empirical panel-data study (abstract)
- Population / context
- Older adults surveyed in the China Longitudinal Aging Social Survey, 2020–2023.
- Summary based on
- Study abstract
- Publication status
- Preprint · peer-review status not confirmed
- Original publication
- September 30, 2026
Older adults surveyed in two waves, 2020–2023
The study asks whether generative artificial intelligence (AI that generates text or other content) affects sports participation among older adults. It uses two waves of panel data (the same people surveyed repeatedly) from the China Longitudinal Aging Social Survey, spanning 2020–2023, and draws on search-as-learning theory. The authors frame the work against population aging and digital transformation and aim to give government agencies policy recommendations on digital tools for proactive health management and sustained exercise.
Supporting source [1]
Link to participation and to risk perception bias
The findings indicate that generative AI significantly promotes sports participation among older adults, and this result held after a battery of robustness checks. Mechanism analysis suggests generative AI fosters participation by reducing health risk perception bias. Compared with bias based on peer comparison, generative AI had a stronger corrective effect on bias based on self-comparison. Extended analyses report significant heterogeneity by prior knowledge level, task type and exercise motivation. These are associations from observational survey data, not proof that generative AI causes the changes.
Supporting source [1]
Abstract-only scope of the AI and exercise study
Only the abstract was reviewed; the full text was not reviewed. The abstract reports the survey source, the two waves and the 2020–2023 span, but gives no sample size, no exact effect sizes and no detailed description of how generative AI use was measured. The mechanisms and the heterogeneity findings are therefore reported at abstract level only, and the policy recommendations are the authors' aims rather than tested interventions.
Supporting source [1]
How to interpret this
Abstract only; the full text was not reviewed. No sample size, effect sizes or measurement details of generative AI use are reported in the abstract. Observational associations, not causal proof.
Based on the study abstract; the full paper, tables and supplementary data were not reviewed.
Source & supporting passages
Yu Z, Li H, Hao J, Chen G. · europepmc.org · September 30, 2026