Artificial intelligence-driven phage therapy in veterinary medicine: an adaptive One Health strategy to mitigate antimicrobial resistance in livestock systems

Antimicrobial resistance (AMR) in animal production systems is a major structural driver of the global resistance crisis. Food-producing animals account for the majority of global antimicrobial consumption, generating sustained selective pressure across livestock, environmental, and zoonotic bacterial reservoirs. Intensive poultry, swine, cattle, and aquaculture systems amplify pathogen transmission and accelerate resistance emergence. Bacteriophage therapy offers a species-specific, microbiome-preserving alternative to conventional antibiotics; however, large-scale veterinary implementation has historically been constrained by challenges including strain-level host prediction, resistance evolution, biosafety considerations, manufacturing scalability, economic feasibility, and regulatory adaptation. Recent advances in artificial intelligence (AI) show promise for enabling precision veterinary phage therapy, though most applications remain at the computational proof-of-concept or preclinical stage. Deep learning and graph-based genomic models have demonstrated high accuracy on benchmark datasets, reinforcement learning has been explored in computational models for cocktail optimization, and AI-assisted genomic screening can enhance biosafety assessment. Integration with real-time AMR surveillance could potentially facilitate adaptive deployment strategies, subject to field validation. Economic modeling suggests that moderate reductions in metaphylactic antibiotic use could yield production and public health benefits, though these estimates remain illustrative. This review synthesizes current evidence on AI-guided phage discovery, epidemiological modeling, microbiome modulation, horizontal gene transfer risk assessment, economic evaluation, and regulatory innovation. Within a One Health framework, adaptive AI-guided phage platforms represent a high-leverage strategy for reducing antimicrobial dependence, provided that critical knowledge gaps are addressed.