Researchers released two AI‑driven advances: DeepTarget, a computational platform that predicts anti‑cancer mechanisms of small molecules and challenges single‑target drug paradigms; and CarbaDetector, a machine‑learning model that identifies carbapenemase‑producing Enterobacterales from disk diffusion tests with high accuracy. Both tools target long‑standing bottlenecks in discovery and diagnostics. DeepTarget, reported by Sanford Burnham Prebys and collaborators in npj Precision Oncology, uses deep computational models to infer likely mechanisms of action from chemical and phenotypic data, enabling faster hypothesis generation for drug developers. CarbaDetector addresses clinical microbiology’s need for rapid, reliable detection of carbapenemase producers and could speed infection control decisions in hospitals. These innovations illustrate how generative and predictive AI are being embedded across the R&D and clinical workflows — accelerating candidate triage and improving laboratory turnaround — while raising questions about validation, deployment, and regulatory acceptance.