From code to cure: Harnessing AI for antibiotic discovery and design

This is the authors' abstract. We don't add key points for this paper.

Artificial intelligence (AI) now enables the exploration of vast chemical and sequence spaces, the design of synthesizable antibiotics, and the prioritization of safe, mechanism-diverse candidates through integrated modeling and accelerated testing. Here, we survey recent advances in AI for both small-molecule and peptide antimicrobial drugs, highlight emerging data and reporting standards, and outline translational pathways and regulatory hurdles needed to bring these agents to patients.