A new AI model, FAHR-FaceSurvival, can predict cancer patient mortality risk using only facial photographs. This tool was developed by researchers at Mass General Brigham and trained on millions of general facial images and thousands of cancer patient images linked to medical outcomes. The AI identified specific facial regions, primarily below the eyes and around the nose bridge, as indicators of survival risk. A higher score from FAHR-FaceSurvival was significantly associated with an increased risk of early mortality. For instance, patients with the highest scores were much more likely to die within three months compared to those with the lowest scores. This research builds upon previous work by the same team, including FAHR-FaceAge, which predicts biological age from facial features. The facial age assessment tool focuses on different facial areas like nasolabial folds and the forehead. Facial aging signs can reflect lifestyle factors such as smoking, alcohol, and drug use, with smoking showing a strong correlation with appearing older. The ultimate goal is to provide physicians with more quantitative and scalable health assessments from facial photos to aid in personalized treatment decisions. This technology aims to estimate physical reserve and frailty for more informed treatment recommendations.