Inductive Bio won an award of up to $21 million from ARPA‑H to lead a consortium building next‑generation AI models for drug toxicity prediction (DATAMAP). The industrial‑academic team includes Amgen, Cincinnati Children’s, Baylor College of Medicine and Torch Bio; they will generate data from organoids, ex vivo human tissue systems and microphysiological models to train AI for drug‑induced liver injury and cardiotoxicity. The program aims to replace or reduce animal testing and improve human predictive power in preclinical safety, addressing a leading cause of late‑stage attrition. Inductive will work with the FDA on potential regulatory use cases and contextualize models for real‑world drug development decisions. If successful, DATAMAP could change preclinical safety assessment standards and accelerate safer candidate selection by integrating human‑relevant biological data with machine learning.