Cardiovascular disease is a core part of internal medicine, where clinicians manage patients with multimorbidity, polypharmacy, repeated transitions of care, and a high burden of fragmented information. In this context, the promise of artificial intelligence (AI) is not to replace clinical judgement, but to help clinicians in internal medicine detect relevant signals earlier, stratify risk more appropriately, and support decisions within real-world care pathways [1–4]. Yet enthusiasm should be tempered by experience: in cardiovascular medicine, many AI tools have shown strong retrospective performance, but far fewer have demonstrated robust clinical utility, safe integration into routine workflows, or measurable benefit for patients and care teams [1,2,5].
