Roundtable at IARU-ALH

October 10th, 2025


  1. Ethical and Privacy Dimensions

How can AI systems balance autonomy and privacy with regards to need to train models on aging populations, particularly when monitoring health and behavior in vulnerable individuals?

  1. Algorithmic Bias and Fairness

In what ways might AI models trained on non-representative (e.g. cross-country) health datasets exacerbate inequalities in aging outcomes across society?

  1. Predictive Analytics and Preventive Health

How reliable are current AI models in predicting age-related diseases (e.g., dementia, frailty) given the complex and heterogeneous nature of the aging process? How accurate do you think these models will ever be able to become?

  1. Cognitive Health and Human-AI Interaction

Can AI-based cognitive assistants effectively enhance mental health and reduce cognitive decline in older adults without fostering overreliance or social isolation?

  1. Personalized Aging

To what extent can AI contribute to quantifying “biological age” as opposed to chronological age, and how might such metrics reshape interventions in geroscience?

  1. Socioeconomic Implications

How might -- within China and across the world -- we better integrate AI into aging care systems, labor markets, and intergenerational equity in resource allocation?

  1. Longitudinal Modeling and Causality

What are the methodological challenges in using AI to infer causal relationships between lifestyle factors and aging trajectories in observational datasets?

  1. The Role of Theory

Traditional models in social epidemiology have relied heavily on theory. What is the role of theory moving forward in an age of big data, machine learning, and high performance computing?