Antimicrobial resistance (AMR) is a global public health concern that has developed at an alarming rate due to rapid genetic mutation, horizontal gene transfer, and the misuse of antibiotics. The current traditional methods used for the detection of ARGs and the development of new antibiotics are not only inefficient but also lack accuracy. AI and ML promise to address AMR through advanced computation techniques such as supervised learning, deep learning, reinforcement learning, and natural language processing. AI-based models use data from genomic sequences, microbiome profiles, and clinical records to predict the emergence of outbreaks of AMR, identify resistance determinants, and aid in drug discovery.

The transformer-based AI models show superior performance in analyzing genomic data, predicting bacterial resistance profiles, and unveiling the key ARG relationships. Furthermore, AI facilitates novel antibiotics through the modeling of the structure–activity relationship, Raman spectroscopy, and mass spectrometry. AI-based clinical decision support systems further optimize antibiotic prescription by reducing the misuse and development of resistance. However, in AMR research, AI faces the challenge of ethical issues and data privacy; the bias found in algorithms poses another significant concern. Future work on AI methods, especially combining pre-trained models of genomics, will enable the improved accuracy of predictions and implementation of precision medicine approaches for infection management. Indeed, with growing advancements in AI technology, this integration will provide the foundation on which the strategy of bacterial resistance control will thrive, preserving the efficacy of available antibiotics.