Effective treatment for Alzheimer’s disease (AD) depends on early detection but current methods for measuring “gold standard” biomarkers are expensive, involve procedures that are uncomfortable and difficult for people to tolerate and thus nearly impossible to do at a large scale. The proposed study seeks to administer a protocol of internet/smartphone-linked applications to Framingham Heart Study (FHS) Generation 2 and multi-ethnic OmniGeneration 1 cohorts. The immediate scientific objective is to collect digital measures of AD related behaviors in order to enhance the sensitivity of traditional health metrics. Using advance data science approaches, these digital metrics make possible expanding the concept of diagnostic AD biological biomarkers to include digital ones. Continuous digital monitoring of this slow disease process would also greatly facilitate the construct and feasibility of conducting large scale clinical trials for treatments that might be most effective if administered much earlier in the disease course. Further use of smart technologies will also enable low-cost longitudinal monitoring of health-related behaviors at the population level, where detecting changes so early (e.g., prognostic digital biomarkers) that interventions may prevent disease altogether.