WASHINGTON, June 27, 2026 — The U.S. Defense Advanced Research Projects Agency (DARPA) is offering American manufacturers of metals, ceramics, and composite materials customized artificial intelligence (AI) tools in exchange for access to multi-year production data. The initiative aims to improve the consistency and efficiency of large-scale materials manufacturing while supporting the U.S. defense industrial base.
The Request for Information (RFI), published on June 25, 2026, under designation DARPA-SN-26-81 and titled "Revolutionizing Industrial Scale Materials Processing," invites participation from companies producing more than 10 metric tons (22,046 pounds) annually of cast, rolled, and forged metals, bulk ceramics, and composite materials. Responses must be submitted by July 24, 2026.
DARPA will use the information to shape a future research program focused on reducing variability in industrial materials processing. Participating manufacturers would receive physics-informed AI tools tailored to their production equipment and manufacturing processes without requiring major capital investments or facility upgrades.
Reducing Material Performance Variability
Materials used in military aircraft, missiles, ships, and armored vehicles naturally vary between production batches due to differences in composition, rolling pressure, heat treatment, cooling conditions, and other manufacturing variables. As a result, engineers design components using the minimum guaranteed material properties rather than average or peak performance to ensure reliability.
According to DARPA, this conservative approach leaves significant performance potential unused. The agency noted that many 7000-series aerospace aluminum alloys, commonly used in aircraft structures and missile airframes, can achieve performance levels more than 15 percent above their minimum specification values. Consistently producing materials closer to these higher performance levels could reduce structural weight, improve system efficiency, and enhance the performance of defense platforms.
AI-Based Manufacturing Optimization
DARPA said large-scale materials production remains heavily dependent on the experience of skilled operators, whose knowledge is often difficult to document or transfer. In addition, manufacturing processes involve numerous interconnected variables, making it challenging to identify which factors most directly influence final material properties.
The agency plans to work with materials scientists, process engineers, and data scientists to develop physics-informed computational models using manufacturers' historical production data. These models are intended to identify causal relationships between processing conditions and material performance, enabling manufacturers to improve production consistency, increase yield, reduce scrap, and minimize rework while using their existing production facilities.
Protecting Proprietary Data
DARPA acknowledged that the requested production data represents valuable proprietary information. To address industry concerns, the agency said future program teams are expected to use secure data enclaves, non-disclosure agreements, and other security measures to protect participating companies' information.
Strengthening Domestic Manufacturing
The initiative supports broader efforts to strengthen the U.S. defense industrial base by improving the efficiency and predictability of domestic production of aerospace-grade metals, ceramics, and composite materials. Rather than requiring new manufacturing facilities, DARPA's approach focuses on helping producers maximize the performance of their existing equipment through advanced AI and data-driven process optimization.
The agency said the resulting technologies could also provide benefits beyond defense manufacturing, supporting more efficient production across commercial aerospace, automotive, energy, and other advanced industrial sectors.
Manufacturers interested in participating have until July 24, 2026, to submit responses that will help define the scope and direction of DARPA's future research program.
——— End of Article ———
