DARPA and the U.S. National Science Foundation, working in collaboration with the Center for Artificial Intelligence Standards and Innovation at the National Institute of Standards and Technology, have developed Artificial Intelligence Forge, a research and development program designed to catalyze breakthroughs in Artificial Intelligence for national security. The initiative is aimed at closing the gap between rapid commercial Artificial Intelligence progress and the distinct requirements of defense and intelligence missions. Its goals include accelerating progress toward Artificial Intelligence that is more reliable and predictable in high-stakes settings, more understandable to operators, and more secure in contested environments, while also building a durable research ecosystem that connects universities, frontier Artificial Intelligence companies, and government.
The program brought together representatives from frontier Artificial Intelligence companies, chief Artificial Intelligence officers from more than 15 Department of War and Intelligence Community agencies, and other government stakeholders to identify core challenges for national security. Their work is captured in the AI Forge Critical Artificial Intelligence Challenges for National Security report, which outlines 15 research challenges in three thrust areas. These areas are Artificial Intelligence interpretability, focused on making system behavior, decisions, and impacts understandable to humans in operational settings; Artificial Intelligence control, focused on tools that can provide strong, verifiable evidence of bounded, auditable, and reliable model behavior while preserving meaningful human control; and adversarial robustness, focused on scientific foundations for systems that maintain integrity and intended performance under deliberate attack. To keep pace with the field, the challenges will be revisited every six months during the program.
Artificial Intelligence Forge is now seeking input from the university research community through a Request for Information tied to the report’s challenge areas. Responses will help create a repository of U.S. universities interested in advancing next-generation Artificial Intelligence research for national security. Responses are due by June 22. The program plans to establish a forum that includes universities, industry, and U.S. government representatives to fund, guide, and manage fast-paced, university-led research projects. The forum is intended to pair academic talent with frontier-scale compute, models, and expertise around mission-driven problems shaped by leaders across the Department of War and Intelligence Community. The forum will be administered by a nonprofit and will launch in summer 2026.