- cross-posted to:
- privacy@lemmy.ml
- cross-posted to:
- privacy@lemmy.ml
I know people have mixed opinions on Braxman but I don’t see any huge leaps in logic here tbh… Thoughts?
I know people have mixed opinions on Braxman but I don’t see any huge leaps in logic here tbh… Thoughts?
Physical access like an NPU chip fixed onto your motherboard?
Sure, anything with direct bus access to unencrypted data… that’ll do it
I didn’t mean that. I meant if the hacker has access to the administrator (or just user in case with E2EE messengers) account, they can see and download anything, no matter how encrypted it is. The chips can do stuff as well but idk any proof of that tbh
Sure, side channel leakage if you can run locally.
Honestly, most machines have enough cores, that you could pin a process to a specific core giving it independent cache, and work around a lot of these side channel attacks. So you’re encrypted end to end messenger would get an exclusive core. Kind of like how we do VM pinning nowadays
What?
Bone apple tea. Fixed
Sus