ABSTRACT

Team Atlanta placed 1st in the DARPA AI Cyber Challenge (AIxCC), earning a $4M grand prize in the final round. In this talk, I will introduce the DARPA AIxCC competition and share our technical approaches that led to victory—specifically, how we augmented large language models (LLMs) with traditional software analysis techniques to automatically discover and repair security vulnerabilities in real-world, large-scale open-source projects.

BIO

Taesoo Kim is a Professor in the School of Cybersecurity and Privacy (SCP) and the School of Computer Science (SCS) at Georgia Tech, and VT in Security Research at Microsoft. Prof. Kim leads Team Atlanta, which won 1st place in the DARPA AI Cyber Challenge (AIxCC), a competition to build autonomous cyber reasoning systems that detect and remediate software vulnerabilities in open-source projects, earning a total award of $6 million prize.

He is the recipient of numerous honors, including the NSF CAREER Award (2018), the Internet Defense Prize (2015), and several best paper/artifact awards at premier venues such as SOSP’21, USENIX Security’18, and EuroSys’17. Prof. Kim holds a B.S. from KAIST (2009) and an S.M. (2011) and Ph.D. (2014) from MIT.