SoK: DARPA's AI Cyber Challenge (AIxCC): Competition Design, Architectures, and Lessons Learned

Cen Zhang, Georgia Institute of Technology and Microsoft; Younggi Park, Independent Researcher; Fabian Fleischer, Yu-Fu Fu, and Jiho Kim, Georgia Institute of Technology; Dongkwan Kim, Microsoft; Youngjoon Kim, Georgia Institute of Technology; Qingxiao Xu, Texas A&M University; Andrew Chin, Georgia Institute of Technology; Ze Sheng, Texas A&M University; Hanqing Zhao, Georgia Institute of Technology; Michael Pelican and David J. Musliner, Smart Information Flow Technologies (SIFT); Jeff Huang, Texas A&M University; Jon Silliman, Mikel Mcdaniel, Jefferson Casavant, Isaac Goldthwaite, Nicholas Vidovich, and Matthew Lehman, Kudu Dynamics; Taesoo Kim, Georgia Institute of Technology and Microsoft

DARPA's AI Cyber Challenge (AIxCC, 2023--2025) is the largest competition to date for building fully autonomous cyber reasoning systems (CRSs) that leverage recent advances in AI --- particularly large language models (LLMs) --- to discover and remediate vulnerabilities in real-world open-source software. This paper presents the first systematic analysis of AIxCC. Drawing on design documents, source code, execution traces, and discussions with organizers and competing teams, we examine the competition's structure and key design decisions, characterize the architectural approaches of finalist CRSs, and analyze competition results beyond the final scoreboard. Our analysis reveals the factors that truly drove CRS performance, identifies genuine technical advances achieved by teams, and exposes limitations that remain open for future research. We conclude with lessons for organizing future competitions and broader insights toward deploying autonomous CRSs in practice.

USENIX Security '26 Open Access Sponsored by
King Abdullah University of Science and Technology (KAUST)

Open Access Media

USENIX is committed to Open Access to the research presented at our events. Papers and proceedings are freely available to everyone once the event begins. Any video, audio, and/or slides that are posted after the event are also free and open to everyone. Support USENIX and our commitment to Open Access.

BibTeX
@inproceedings {320845,
author = {Cen Zhang and Younggi Park and Fabian Fleischer and Yu-Fu Fu and Jiho Kim and Dongkwan Kim and Youngjoon Kim and Qingxiao Xu and Andrew Chin and Ze Sheng and Hanqing Zhao and Michael Pelican and David J. Musliner and Jeff Huang and Jon Silliman and Mikel Mcdaniel and Jefferson Casavant and Isaac Goldthwaite and Nicholas Vidovich and Matthew Lehman and Taesoo Kim},
title = {{SoK}: {DARPA{\textquoteright}s} {AI} Cyber Challenge ({{{{{AIxCC)}}}}}: Competition Design, Architectures, and Lessons Learned},
booktitle = {35th USENIX Security Symposium (USENIX Security 26)},
year = {2026},
isbn = {978-1-939133-58-8},
address = {Baltimore, MD},
pages = {2705--2724},
url = {https://www.usenix.org/conference/usenixsecurity26/presentation/zhang-cen},
publisher = {USENIX Association},
month = aug
}