Dmitrii Kriukov, Evgeniy Efimov, Mikhail S. Gelfand, Alexey Moskalev, Ekaterina E. Khrameeva
Aging clocks use machine learning to estimate biological age as a proxy for general health state. Here, we critically examine their practical value, highlighting fundamental challenges: abstract definitions, inconsistent clinical validation, and ignored prediction uncertainty. By comparing aging clocks with expert risk scores, direct outcome predictors, and emerging large health models, we question their benefits and encourage researchers to explicitly justify clock advantage over established alternatives, ensuring truly actionable insights.