Lilly’s Phase 1 base-editing results on VERVE-102 are paired with continued momentum in AI-enabled precision approaches for disease biology and clinical decision-making. Separate disclosures from university and industry research teams underscore how computational tools are increasingly used to connect biomarker patterns or tumor genetics to likely treatment responses. In one example, UC San Diego researchers described MutationProjector, an AI framework trained on more than 30,000 solid-tumor cases across 10 cancer types and reported performance that matched or exceeded existing prediction methods in multiple cohorts. In a different direction, work on early detection and risk stratification—such as machine learning for neonatal sepsis prediction—points to the growing role of models in clinical workflow. Taken together, these releases show biotech’s parallel push on genetic medicine execution (including gene editing) and on interpretable model-building for patient selection—both areas that can shorten development cycles if they translate into validated clinical utility.