DARPA’s I2O is seeking proposals for unconventional computational approaches that are outside the mainstream, challenge accepted assumptions and have the potential to radically change established practices, and the state of the art, in support of the following thrust areas. Proposed research should investigate innovative approaches that enable revolutionary advances in software science, technology, or systems.
- Proficient Artificial Intelligence – new directions for research that will result in safe, secure and trustworthy systems for national security missions. Of particular interest is accelerating the pace of science and engineering for national security applications.
- Resilient, Adaptable and Secure Systems – enhancing the security and resilience of both new and legacy software with a focus on practical and scalable approaches. Emphasis on formal methods development while leveraging AI and machine learning.
- Advantage in Cyber Operations – leveraging and advancing state-of-the-art AI and secure resilient technologies to produce trustworthy cyber tools and capabilities that operate beyond the capacity or speed of humans. Efforts in the thrust anticipate adversary countermeasures to create enduring capabilities.
- Confidence in the Information Domain – protecting, detecting attacks on, and measuring the health of the information domain, broadly construed. Spans the cognitive level of beliefs and attitudes; the knowledge level of scientific discourse, the financial system, supply chains etc; the tracking level; and the transport level.
Submissions outside the thrust areas may be considered if they involve the development of novel capabilities having a promise to provide decisive information advantage.
Abstracts (encouraged but not mandatory) will be accepted until 1 October 2025.
Full proposals will be accepted and assessed on a rolling basis until 29 October 2025.
Document
Broad Agency Announcement – I2O Office-Wide: HR001125S0002
