IEEE Spectrum: “The Defense Innovation Unit (DIU), part of the U.S. Department of Defense (DOD), is leading an experimental project, Thunderforge, to build a custom agentic AI system with multiple digital “agents” critiquing war plans across different military domains, running parallel analyses, and flagging potential weaknesses neglected by human planners. The U.S. Indo-Pacific Command (INDOPACOM) tested some of Thunderforge’s AI capabilities during a tabletop exercise in June. The project, first announced in March, remains in the early stages, with an initial rollout targeting commanders and planners across INDOPACOM and its sister command in Europe (EUCOM). Ultimately, the system will query internal databases, run DOD-grade simulations, and integrate with pre-existing software, such as DARPA’s SAFE-SiM modeling and simulation architecture, to create vast numbers of realistic and plausible military scenarios for study. California-based Scale AI is steering the project, with Microsoft supplying its large language model (LLM) technology and Anduril providing modeling. Scale AI says its system is designed to coordinate multiple custom agents, each leveraging a range of models and acting as a digital staff officer to help synthesize data for mission-critical planning activities. “These agents dynamically collaborate, fusing separate analyses into a more comprehensive view for operational planners to consider,” says Dan Tadross, the head of public sector at Scale AI. “This approach is designed to shift the role of the operator from being ‘in the loop’—micromanaging a single process—to being ‘on the loop,’ where they can apply their strategic judgment to the options generated.”…