A collection of articles on application of AI for Vulnerability Research
What if we could determine the performance of a fuzzer by just looking at its compiled source code? It turns out we can!
There is a lot of code reuse in software,
Because of Fuzz blockers fuzzing may stop sooner than needed, delaying vulnerability detection, in some cases even for years to come.
Fuzz4All is useful for generating fuzzing code in multiple programming languages. FuzzCoder augments American Fuzzing Loop with LLM based input mutations.
