Re:How to Break into Security, Ptacek Edition
by ronsor on Tuesday November 17, 2026 @12:13AM
(#50527715)
FHE is still really cool, if the performance issues can be solved.
Re:How to Break into Security, Ptacek Edition
by cyberax on Tuesday November 17, 2026 @12:22AM
(#50527768)
It's cool, along with ZKPs but I struggle to find a useful niche for them. Age/ID verification is cool and all, but this is apparently it?
(And I'm ignoring all the blockchain nonsense on purpose)
Algorithms are a solved problem. Post-quantum encryption was the last bit of excitement for a while. I guess ZKPs and FHE are still available, but they are really niche.
It's everything else that fills me with dread: we're using oodles of software written in C/C++, so we're one missing array boundary check away from the global economy collapsing.
Then there's phishing and social engineering. If one of our developers gets compromised, it's trivial to push malicious code into our repo. And from there it's an easy escalation into our customers' billing, private information, etc. We store the bare minimum of data possible, but it's still a lot.