The escalating threat of antimicrobial resistance has renewed global interest in peptide-based antibiotics as adaptable and effective alternatives to conventional small molecules. Peptides possess diverse mechanisms of action, high target specificity, and structural flexibility, which collectively limit the emergence of resistance. This review outlines recent advances spanning the discovery, optimization, and application of peptide antibiotics, from their biological origins and structural classifications to emerging strategies involving artificial intelligence, synthetic biology, and modern delivery technologies. Peptide antibiotics can be categorized by origin as natural, semi-synthetic, or fully synthetic, and further organized by structural class such as α-helical, β-sheet, cyclic, and extended forms. They are also grouped by function into membrane-targeted and non-membrane-targeted types. These classification schemes are not only descriptive but also critical for understanding the therapeutic potential of peptides, as each category presents distinct advantages and engineering challenges that influence stability, specificity, and overall clinical performance. Advances in artificial intelligence, synthetic biology, and continuous manufacturing are reshaping how peptide drugs are designed and produced, while innovations in drug delivery systems are addressing critical issues of stability and bioavailability. Together, these developments are laying the foundation for a new generation of peptide-based therapeutics capable of meeting the evolving challenges of antimicrobial resistance.

Two Gram-stain-negative, aerobic, non-motile, rod-shaped bacterial strains, designated IMCC43444^T and IMCC44478^T, were isolated from surface seawater collected off Deokjeok Island and Jangbong Island, respectively, in the Yellow Sea. The two strains shared 100% 16S rRNA gene sequence similarity with each other but exhibited ≤ 96.2% similarity to validly published species of the genus Robiginitalea. Complete whole-genome sequences of IMCC43444^T and IMCC44478^T were 3.21 Mb and 3.30 Mb in size, with DNA G + C contents of 46.5% and 46.4%, respectively. Genome-based relatedness analyses revealed average nucleotide identity (ANI) and digital DNA–DNA hybridization (dDDH) values of 90.7% and 42.9% between the two strains, which are well below the accepted species-level thresholds. Furthermore, ANI (≤ 70.2%) and dDDH (≤ 17.8%) values relative to type strains of Robiginitalea species supported the conclusion that strains IMCC43444^T and IMCC44478^T each represent novel species within the genus. Chemotaxonomic characterization showed that iso-C_15:0, iso-C_17:0 3-OH and iso-C_15:1 G were the major fatty acids of both strains; menaquinone-6 (MK-6) was the sole isoprenoid quinone; and the major polar lipids comprised phosphatidylethanolamine, glycolipids, aminolipids, phospholipids, and other unidentified lipids. Based on phylogenetic, genomic, and phenotypic evidence, strains IMCC43444^T and IMCC44478^T are proposed as two novel species, Robiginitalea rubriflava sp. nov. and Robiginitalea insularis sp. nov., respectively. The type strains are IMCC43444^T (= KCTC 102397^T = JCM 37893^T) and IMCC44478^T (= KCTC 102398^T = JCM 37894^T).