So, the problem with USB tokens that we basically have two choices: - Unauditable black boxes built on *supposedly* more secure ICs that require NDAs to develop for - Open and auditable, but definitely pwnable off the shelf microcontrollers. Which poison do you prefer?
-
Show this thread
-
So far I've favored the first one, because you *know* all those keys based on a random STM32 are going to be glitchable, and closed stuff *might* be better. But after ROCA and other fails, maybe open firmware is better and just hope nobody physically attacks your key?
3 replies 0 retweets 10 likesShow this thread -
Keep in mind that this depends on your use case, e.g. a key keeping SSH keys needs to be more secure than a key used for FIDO U2F login for a handful of websites, because the latter is a second factor only and easier to revoke/replace.
2 replies 0 retweets 10 likesShow this thread -
Also keep in mind that open does not mean secure, I've seen some absolutely abysmal firmware in "open" firmware projects along these lines too. I guess what I'm asking is whether it's worth doing #2 "right" with the caveat of being vulnerable to physical attacks.
3 replies 0 retweets 8 likesShow this thread -
Replying to @marcan42
Are you sure #1 is not vulnerable to physical attacks?
1 reply 0 retweets 0 likes -
Replying to @SwissHttp
No, it just ostensibly tries harder than #2. I honestly have no idea why absolutely no secure element vendor has tried opening up their docs. It's ridiculous.
1 reply 0 retweets 0 likes -
Replying to @marcan42
That might be a business case. Or, maybe if you see the NDA‘d docs, you immediately know it’s vulnerable by design.
1 reply 0 retweets 0 likes
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.
