Two advances aim to sharpen antimicrobial resistance (AMR) detection and prediction. Ma and Kim introduced the dilution‑and‑delay (DnD) susceptibility assay in Nature Communications, a high‑throughput method that reveals hidden resistance phenotypes missed by standard testing. The assay produces high‑resolution growth dynamics enabling more sensitive detection of evolving resistance. Separately, AMR‑GNN, a graph neural network framework, demonstrated improved genomic prediction of resistance by integrating multi‑representation genomic features, offering a computational route to forecast resistance from sequence data. Both tools address urgent gaps in clinical and surveillance capability for AMR. Deployment of DnD in clinical labs and integration of AMR‑GNN into genomic surveillance pipelines could accelerate actionable detection of resistance and inform antibiotic stewardship and drug‑development priorities.