César de la Fuente’s group at the University of Pennsylvania is using artificial intelligence to scour genomes, venoms and ancient sequences for antimicrobial peptides, reporting successive discoveries including candidates from archaea and animal venoms and proposing molecular de‑extinction screens for extinct species. The team’s work and warnings about a thin antibiotic pipeline were highlighted in a Physical Review Letters essay and coverage summarizing de la Fuente’s lab strategy. The group trains AI to predict peptide sequences with antimicrobial activity and then synthesizes candidates for experimental testing. The approach expands the searchable chemical space and has yielded promising leads that could address drug‑resistant bacteria. Antimicrobial peptides are short amino acid chains that can disrupt microbial membranes or processes; AI accelerates their discovery by predicting structure–activity relationships across large sequence databases.