This study introduces a deep-learning model, CXR CVD-Risk, which estimates the 10-year risk for major adverse cardiovascular events (MACE) from a single chest radiograph (CXR). The model was developed using data from the Prostate, Lung, Colorectal, and Ovarian (PLCO) Cancer Screening Trial and externally validated in two cohorts of outpatients potentially eligible for primary cardiovascular prevention. The first cohort included 8869 outpatients with unknown ASCVD risk due to missing inputs, and the second cohort included 2132 outpatients with known ASCVD risk. The CXR CVD-Risk model showed higher net benefit than baseline strategies in the cohort with unknown ASCVD risk and had similar performance and additive value to the traditional ASCVD risk score in the cohort with known ASCVD risk. In the cohort with unknown ASCVD risk, those identified as statin-eligible by CXR CVD-Risk had a 1.5-fold higher 10-year risk for MACE compared to those classified as ineligible, independent of available baseline cardiovascular risk factors. These findings suggest that CXR CVD-Risk could enable population-based opportunistic screening using routine CXRs to identify individuals at high risk for cardiovascular disease, prompting risk factor assessment and targeted prevention.This study introduces a deep-learning model, CXR CVD-Risk, which estimates the 10-year risk for major adverse cardiovascular events (MACE) from a single chest radiograph (CXR). The model was developed using data from the Prostate, Lung, Colorectal, and Ovarian (PLCO) Cancer Screening Trial and externally validated in two cohorts of outpatients potentially eligible for primary cardiovascular prevention. The first cohort included 8869 outpatients with unknown ASCVD risk due to missing inputs, and the second cohort included 2132 outpatients with known ASCVD risk. The CXR CVD-Risk model showed higher net benefit than baseline strategies in the cohort with unknown ASCVD risk and had similar performance and additive value to the traditional ASCVD risk score in the cohort with known ASCVD risk. In the cohort with unknown ASCVD risk, those identified as statin-eligible by CXR CVD-Risk had a 1.5-fold higher 10-year risk for MACE compared to those classified as ineligible, independent of available baseline cardiovascular risk factors. These findings suggest that CXR CVD-Risk could enable population-based opportunistic screening using routine CXRs to identify individuals at high risk for cardiovascular disease, prompting risk factor assessment and targeted prevention.