Artificial Intelligence in genetic healthcare: toward adaptive and systems-level genomic medicine

Artificial intelligence is reshaping genetic healthcare through increasingly sophisticated approaches to variant interpretation, disease prediction, biomarker discovery, gene editing, and precision therapeutics. Yet these applications remain fragmented, often treating genomic and clinical information as static observations rather than components of biological systems that evolve during disease. This review connects current AI capabilities with a next-generation model of adaptive genomic medicine integrating longitudinal multi-omics, clinical and environmental data, explainable AI, federated learning, network-level biomarkers, and continuous outcome feedback. We extend this framework beyond conventional genomic prediction by considering pathway interactions, compensatory biological redundancy, network reorganization, and cross-system changes as potentially informative features of disease progression. We further propose longitudinal approaches for tracking changing biological relationships and resilience before conventional biomarkers become abnormal. Together, these concepts reposition genomic AI from a collection of predictive tools toward a dynamic systems-level framework capable of learning how disease evolves, anticipating state transitions, and supporting increasingly individualized clinical intervention.