Antimicrobial resistance (AMR) poses a threat to global health. One of the greatest challenges in AMR is the shortcoming in diagnosis. Classic approaches to identify multidrug-resistant bacteria and fungi are based on relatively slow microbiological assays, while molecular approaches require prior knowledge on drug resistance genes. Here we propose to develop a fast RNA based method to study cellular fitness and predict antimicrobial resistance. We will pilot the use of RNA metabolism signatures for the fast identification of AMR in infectious diseases. We will benchmark traditional culture-based tests with a fast molecular readout of cellular fitness and an AI-based classifier to identify antimicrobial phenotypic susceptibility.