This chapter looks at how generative artificial intelligence and graph neural networks are changing drug discovery for antimicrobial and antiviral resistance. It covers the global challenges from ESKAPE bacteria, multidrug-resistant tuberculosis, and resistant viruses like HIV and SARS-CoV-2, and explains why flexible AI methods are important. The chapter introduces key generative models, such as variational autoencoders, generative adversarial networks, diffusion models, transformers, and different types of graph neural networks. It explains how these models help represent molecules, design new compounds, and predict several properties at once. The chapter also reviews main data sources, including genomic, structural, and chemical databases, and discusses the difficulties in combining these data to create strong training sets. It describes full generative pipelines for finding antibacterial and antiviral agents, from choosing targets and generating molecules to optimizing and testing them in the lab. Case studies, like AI-designed antibiotics such as abaucin and inhibitors of the SARS-CoV-2 main protease, show how using AI with experiments can lead to promising drug candidates. The chapter also points out the need for good data, ongoing resistance monitoring, and global teamwork.