June 1, 2026
Commercial artificial intelligence (AI) development is progressing at a breathtaking pace. Yet many of the most consequential AI challenges for national security remain underexplored because they lack immediate commercial applications and are not the primary focus of private industry. There is a critical need to bridge the gap between commercial AI innovation and the unique requirements of national security.
To this end, DARPA and the U.S. National Science Foundation (NSF) have jointly developed a research and development program called AI Forge to catalyze breakthroughs in AI for national security, working in close collaboration with the Center for AI Standards and Innovation (CAISI) at the National Institute of Standards and Technology. AI Forge aims to accelerate progress towards AI that is significantly more reliable and predictable in high-stakes settings, understandable to its operators, and secure in contested environments. It also envisions building a durable research ecosystem around priority AI challenges and enabling a more robust exchange of talent and ideas across universities, frontier AI companies, and government than is possible today. AI Forge is strategically aligned with America’s AI Action Plan.
The program convened representatives from frontier AI companies, chief AI officers from more than 15 Department of War (DOW) and Intelligence Community (IC) agencies, and other government stakeholders to explore and reach consensus around core AI challenges for national security. These challenges comprise the newly released AI Forge Critical AI Challenges for National Security report, which will serve as a roadmap to focus research under the program.
The report synthesizes insights from experts across industry and government into 15 research challenges in three thrust areas for which university-led teams will develop aligned ideas and research proposals. The thrust areas are:
- AI interpretability: Research challenges focused on making the behavior, decisions, and impacts of AI systems understandable to humans, with an objective to move beyond explanations in routine settings toward operational interpretability.
- AI control: Research challenges focused on pioneering tools that can provide strong, verifiable evidence of bounded, auditable, and reliable model behavior today, while laying the essential groundwork for maintaining meaningful human control over future, more capable AI systems.
- Adversarial robustness: Research challenges focused on building the scientific foundations for AI that is not just capable, but resilient by design so that it maintains its integrity and intended performance even when under deliberate attack from a thinking adversary.
The program hypothesizes that pre-competitive AI research in these thrust areas can accelerate the adoption of AI innovations by industry and federal agencies. To reflect the fast-changing technical AI research landscape, the challenges will be revisited every six months during the program.
A call for bold thinking
AI Forge is calling upon the university research community to share their capabilities to conduct research on the challenges described in the program’s Critical AI Challenges for National Security report. University researchers interested in submitting their capabilities are encouraged to do so through the AI Forge Request for Information. Responses to this RFI will be used by the program stakeholders to establish a repository of U.S. universities interested in accelerating next-generation AI research to solve national security challenges. Responses are due by June 22.
A forum for unlocking national-security focused AI innovation
Moving forward, the program will establish a forum comprised of universities, industry, and U.S. government representatives to fund, guide, and manage fast-paced, university-led research projects. The forum aims to combine academic talent with frontier-scale compute, models, and expertise to address mission-driven challenges informed by national security AI leaders from across the DOW and IC.
“We’re taking a unified approach to create breakthroughs in AI for national security,” said Matthew Marge, DARPA program manager for AI Forge. “The frontier AI companies build and commercialize massive, high-capability models and compute. Universities are engines of deep, foundational research and, importantly, they cultivate our nation's future talent. At the intersection is high-risk, high-reward research that requires both massive scale and deep, mission-driven work – something that is difficult to pursue in either environment alone. With AI Forge, we’re looking to build an ambitious new ecosystem that bridges this gap.”
“NSF is excited to partner with DARPA, working alongside CAISI, on this groundbreaking effort to catalyze AI innovation for national security” said Erwin Gianchandani, NSF Assistant Director for Technology, Innovation and Partnerships. “By linking the rapid advances of frontier AI companies, the research and talent at universities, and the use cases surfaced by the intelligence community, AI Forge will propel advancements in AI capabilities for the benefit of U.S. national security and, ultimately, all Americans.”
The forum will be administered by a nonprofit and will launch in summer 2026. More details will be available in the coming weeks.
###
Media with inquiries should contact DARPA Public Affairs.
