The Boston-based company is developing toxicology-specific models and AI agents to help scientists connect evidence across preclinical safety studies, building on research from Mass General Brigham.
BOSTON--(BUSINESS WIRE)--Tremont AI, a biomedical AI company focused on preclinical drug safety, today emerged from stealth. The company is developing multimodal foundation models and AI agents to help scientists analyze and interpret evidence across preclinical safety studies.
Drug discovery is generating new candidate compounds faster than ever, in part due to AI. But before any of them can reach patients, toxicologic pathologists specializing in drug safety assessment must manually examine huge amounts of tissue data to characterize a compound's toxicity. As drug development accelerates, that process is becoming increasingly difficult to scale.
Tremont AI's foundation models build on technology exclusively licensed from Mass General Brigham. Tremont AI is building toxicology-specific foundation models designed to help pathologists and toxicologists work with this information more effectively. TRACE analyzes pathology images, ToxScribe turns what the models see into language, and Tremont Studio brings these capabilities together so evidence from across a study can be reviewed and interpreted in context across multiple modalities.
“Preclinical safety is a complex problem built on enormous amounts of biological data,” said Luca Weishaupt, Chief Executive Officer and co-founder of Tremont AI. “We are building Tremont to give scientists a new way to work with that complexity and help connect evidence across entire tox studies.”
“Foundation models and AI agents have the potential to accelerate toxicology workflows and assist toxicologic pathologists in analyzing and interpreting complex preclinical data,” said Faisal Mahmood, PhD, scientific co-founder of Tremont AI, Associate Professor of Pathology at Brigham and Women's Hospital and inaugural Director for Mass General Brigham's AI Institute. “These approaches could help integrate evidence across modalities and studies, support more systematic characterization of biological effects and potential toxicities, and strengthen the scientific basis for drug candidate assessment.”
Tremont AI is launching as preclinical safety assessment enters a period of rapid modernization. Regulatory agencies and drug developers are increasingly exploring computational methods alongside traditional approaches, creating growing demand for technologies that can make existing safety data more quantitative, reproducible, and useful.
Contacts
Media Contact: Luca Weishaupt - media@tremont.ai
