A key factor to optimal acceptance and comfort of automated vehicle features is the driving style of the automation. Mismatches between the automated vehicle and the driver-preferred driving styles can make drivers take over more frequently or even disable the automation features. We propose a framework that adaptively changes the driving style of an SAE L2 driving automation to match the driver’s preference. We conduct a driving simulator study to have participants interact with L2 driving automation with different driving style adaptations. We analyze the effects of different AV driving style adaptation heuristics on drivers’ preferences and trust. We then develop a driving preference prediction model to identify the change in preferred driving styles. Using this model, we develop and validate an implicit adaptation of the driving style algorithm to minimize the driving style preference mismatch. The validation study show that the proposed adaptive algorithm obtain an equal or lower mismatch in preference as well as higher average trust in driving automation. The results provide a step toward developing human-aware driving automation that can implicitly adapt its driving style based on the driver’s preference.