A foundation AI model trained on over 10 million ECG recordings successfully identified cardiovascular disease risk by measuring how much a patient's heart electrical patterns deviated from their chronological age. In external validation across 160,493 hospital patients, those with positive 'age acceleration' showed significantly higher all-cause mortality risk, with the strongest predictive power in adults under 65. The model worked across multiple ECG lead configurations, including single-lead setups suitable for wearable devices. This represents a potentially transformative development in cardiovascular risk assessment, offering a non-invasive biomarker that could democratize heart disease screening through consumer devices. The approach builds on emerging biological age research, where discrepancies between chronological and physiological age predict health outcomes. However, the researchers identified morphological confounds from conditions like left bundle branch block, leading them to propose absolute age deviation as more robust. As an unreviewed preprint, these promising findings require peer review and prospective clinical validation before implementation. The work suggests ECG biological age could become as routine as blood pressure measurement for cardiovascular risk stratification.