←
Ivan Fratric 💙💛1,277 posts
Ivan Fratric 💙💛 avatar

Ivan Fratric 💙💛

@ifsecure

Tech lead and security researcher at Google Project Zero. Author: Jackalope, TinyInst, WinAFL, Domato. PhD. Tweets are my own. Backup @ifsecure@infosec.exchange

◷ Joined August 2011▧ 30 Photos and videos
Samuel Groß avatar
⟳ Ivan Fratric 💙💛 retweeted
Samuel Groß@5aelo·Oct 29

After a few years in beta, the V8 Sandbox is now considered a real security boundary! (Just announced at OffensiveCon Tokyo together with @mlippautz). This is less about reaching a particular software milestone than about two things:

Post attachment
Ivan Fratric 💙💛 avatar
Ivan Fratric 💙💛@ifsecure·Oct 29

To expand on this a bit: Yes, the LLM found more bugs than the fuzzer. But the fuzzer finding bugs the LLM didn't shouldn't come as a surprise to anyone who has ever performed a thorough manual review, only to run a fuzzer afterward and discover issues they missed.

People often ask me how I think about fuzzers and other tooling in the age of AI. Instead of my opinion, here is some hard data by @5aelo.

Ivan Fratric 💙💛 avatar
Ivan Fratric 💙💛@ifsecure·Aug 25

The latest TinyInst includes a quick fix for that. It worked in my experiments, but if you encounter any other issues, please report them. Good for me that there is only one place in the TinyInst codebase from which all target process memory is read and written :) (2/2)

Ivan Fratric 💙💛 avatar
Ivan Fratric 💙💛@ifsecure·Aug 25

At this year's OffensiveCon, @R00tkitSMM quite correctly pointed out to me that, on Apple devices, MTE can be used to get more reliable crashes while fuzzing. He also pointed out that Jackalope/TinyInst does not support binaries with MTE enabled and crashes. (1/2)