AI powered future of automated vulnerability fixes

As AI continues to advance at a rapid pace, so does its ability to uncover hidden security vulnerabilities in all types of software. Every bug uncovered presents an opportunity to patch and strengthen code. However, as detection methods improve, it's crucial to be prepared with new automated solutions that enhance our ability to address these bugs.

This talk shares lessons from the experience of Machine Learning Engineers leveraging AI to scale bug-fixing capabilities, particularly focusing on bugs identified by sanitizers in C/C++, Java, and Go code at Google.

It also provides background on the importance of this endeavor, how others can apply these lessons, and highlights other notable work in this area