Artificial intelligence for early detection and risk prediction of antimicrobial resistance in aquatic ecosystems

Aquatic environments are key reservoirs and dissemination pathways of antimicrobial resistance (AMR). However, current water-based surveillance remains fragmented and inefficient for the timely detection of emerging threats. Integrating artificial intelligence with embedded metadata provides a powerful pathway to identify novel antimicrobial resistance genes, characterize resistome profiles, and predict AMR dynamics in real-time by combining omics, environmental, and hydrological data into spatiotemporal predictive models. Successful implementation of this framework will require robust governance, ethical safeguards, and capacity building to support predictive AMR monitoring aligned with the One Health approach.