Abstract: Predictions guide important adaptation responses--from treating patients in hospitals to pretreating roads before snow storm. Advances in machine learning and artificial intelligence are accelerating improvements in prediction accuracy. However, it is unclear how prediction improvements should shape optimal adaptation. I develop a theoretical model for prediction-based prevention and provide three key insights. First, better predictions lead to more intense, yet less frequent, adaptation response. Second, risk preferences matter less as improved predictions resolve more uncertainty. Third, the average adaptation declines for highly risk-averse decision-makers but may rise for less risk-averse ones. These findings highlight the need to align adaptation planning with prediction skill, especially given varying levels of trust in prediction technologies.