Despite advances in cardiology, predicting heart attacks remains a challenge. Now, AI is being applied to this problem, with startups like Bunkerhill Health, Nanox.AI, and HeartLung Technologies using algorithms to screen millions of CT scans for early indicators of heart disease. This approach leverages existing chest CT scans, often performed for other reasons, to identify coronary artery calcium (CAC), a marker for heart attack risk that may be overlooked in standard radiology reports. Dedicated CAC testing is currently underutilized. AI-driven analysis could significantly expand access to this metric by calculating CAC scores from routine chest CTs. These algorithms could then alert patients and doctors to elevated scores, prompting further evaluation and intervention. While still in its early stages, this technology has the potential to identify high-risk patients who might otherwise be missed. More expert groups are endorsing CAC scores to refine cardiovascular risk estimates and persuade skeptical patients to start taking statins. However, the widespread adoption of AI-derived CAC scores raises important questions. A 2022 Danish study showed no mortality benefit from population-based CAC screening, suggesting that universal screening may not be effective. The healthcare system also needs to be prepared to act on the increased number of abnormal findings that AI will generate. Concerns exist about potentially unnecessary procedures and costs, especially since AI-derived CAC scoring is not yet widely reimbursed. Beyond practical considerations, AI-driven diagnosis could fundamentally change how we define disease. As Adam Rodman of Beth Israel Deaconess Medical Center points out, this approach shares similarities with the discovery of 'incidentalomas' on CT scans, disrupting traditional diagnostic pathways. We may be entering an era of 'machine-based nosology,' where algorithms define diseases on their own terms, potentially creating a two-tiered diagnostic system. Ultimately, the effectiveness of AI-derived CAC scores in improving patient outcomes at scale remains to be seen, and clinicians will continue to play a crucial role in interpreting algorithmic outputs and guiding patient care.