In the first episode in the Fuzz Theory series, we work on a program which can be used to auto-generate fuzzing analogues. We explore a few methods of generating these programs, with varying performance properties.
Conversation
The goal here is that if we learn the mathematical properties of fuzzers, and thus, what a fuzzer with a blind spot of a certain type looks like. We could allow a user to know the "coefficients" of their fuzzer, which may help identify the top blind spot it has and fix it!
3
1
10
Replying to
I would love to see a new DARPA program centered around this area of research.