César de la Fuente’s team at the University of Pennsylvania is training machine-learning models to search genomes for novel antimicrobial peptides, uncovering candidates in archaea, animal venoms, and even extinct-species sequences. The approach combines computational screening with experimental validation to build peptide libraries that may counteract drug-resistant bacteria. De la Fuente argues that AI-driven peptide discovery can lower development costs and accelerate lead identification in a field where the antibiotic pipeline has dwindled. Early results from his group include promising sequences recovered from unexpected biological sources; the program aims to translate computational hits into preclinical efficacy against resistant pathogens.