Re:Safe curves for Elliptic Curve Cryptography [pdf]
by tptacek on Wednesday August 14, 2024 @09:27AM
(#41245821)
This is really overthinking it. Dual EC is an asymmetric RNG; it "encrypts" its state to a public key, with an undisclosed private key. To believe it wasn't backdoored (as some people, me included, did --- long story!) you have to take on faith that nobody in the world actually has that private key. We're now fairly sure someone does. That's the whole story.
[ Parent
| on HN ]
Re:Safe curves for Elliptic Curve Cryptography [pdf]
by red_admiral on Wednesday August 14, 2024 @10:12AM
(#41246309)
You're probably right, I'm just saying that if you want to design a RNG that doesn't look extremely suspicious, you do it in such as way that there is no way to have a private key to start with. For example with a hash function somewhere in the loop, since these are (presumably) one-way.