OBJECTIVES: This study aimed to explore a broad range of predictors of generativity in older adults. The study included over 60 predictors across multiple domains, including personality, daily functioning, socioeconomic factors, health status, and mental well-being. METHODS: A random forest machine learning algorithm was used. Data were drawn from the Midlife in the United States (MIDUS) survey. RESULTS: Social potency, openness, social integration, personal growth, and achievement orientation were the strongest predictors of generativity. Notably, many demographic (e.g., income) and health-related variables (e.g., chronic health conditions) were found to be much less predictive. DISCUSSION: This study provides new data-driven insights into the nature of generativity. The findings suggest that generativity is more closely associated with eudaimonic and plasticity-related variables (e.g., personal growth and social potency) rather than hedonic and homeostasis-oriented ones (e.g., life satisfaction and emotional stability). This indicates that generativity is an inherently dynamic construct, driven by a desire for exploration, social contribution, and personal growth.