<p>This article highlights key applications of artificial intelligence (AI) in clinical medicine: <i>Image analysis</i>: AI optimizes cystoscopy for bladder tumors and assists in detecting pneumonia in chest X‑rays with high sensitivity and specificity. It also aids in detecting basal cell carcinomas and enhances detection of diabetic retinopathy using deep learning. <i>AI prediction via genomic analyses</i> (multi-OMICs): AI can already predict therapy response and survival probability, such as the benefit of paclitaxel chemotherapy in gastric cancer or response to neoadjuvant chemotherapy in muscle invasive bladder cancer. <i>AI in real-time monitoring</i>: AI algorithms continuously capture data (e.g., via wearables) for early detection of complications or to build predictive models for therapy response in intensive care units. Clinical decision support systems (CDSS) integrated with AI are essential for improving the quality of evidence-based treatment recommendations and reducing workload. An example project discussed is <i>KITTU</i>, an AI-supported system designed to generate evidence-based treatment recommendations (TR) for prostate cancer, urothelial carcinoma (UC) and renal cell carcinoma (RCC) within multidisciplinary cancer conferences (MCCs). The system converts retrospective patient data and knowledge from clinical studies and guidelines into software-compatible formats to create explainable AI recommendations. KITTU employs a&#xa0;two-step classification: first, superordinate “high-level” recommendations (e.g., ‘surgery’, ‘anti-cancer drug’), followed by more specific “low-level” recommendations (e.g., ‘cystectomy’, ‘pembrolizumab’). The explainability of these recommendations is provided by SHAP (SHapley Additive exPlanations) values, which indicate how specific patient features positively or negatively influenced the prediction. This information is visualized on an interactive dashboard for medical staff.</p>