The Biothreats Emergence, Analysis and Communications Network (BEACON)  announced public availability of PandemIQ Llama, an open-source, domain-trained large language model designed specifically for pandemic surveillance and biothreat analysis. BEACON, based at Boston University’s Center on Emerging Infectious Diseases (CEID),  is a web-based platform and accompanying public health program which aims to address this need by leveraging advanced AI and a global network of human subject matter experts to rapidly collect, analyze, and disseminate information on emerging disease threats.

Developed by data scientists and engineers at the Hariri Institute with input from infectious disease experts from CEID, PandemIQ Llama helps analysts sift through incoming disease signals from sources like HealthMap, public health partners, and global outbreak reporters. These signals are first verified by human outbreak specialists through digital investigation and field communication, then selected by editorial staff for reporting. The model is used to generate initial report drafts, which human experts refine with additional context and analysis before publication—illustrating a hybrid system where AI supports, but does not replace, expert judgment.