I don't love it, but I have to acknowledge that a frontier model building a feature that uses cryptography is going to have a better track record than the median human developer. The immediate practical problem with cryptography engineering has always been that there's a huge amount of detail you need to understand in order to get something right, but only a little bit to know to get something working. LLMs know all the details. They take shortcuts, but fuckups in mainstream cryptography features aren't really about shortcuts, they're about knowledge.

The most important thing to do with cryptography features is to stress test them ("spend $10 on verification for every $1 on implementation"), and that too has become a lot easier with frontier models.

I'm not clear why someone would roll their own crypto. There are so many good solutions out there that are free that it seems pointless.

I keep hearing about post-quantum being an issue though.

It's time to get over the "never roll your own crypto!" bs.

AI can roll you crypto far better than what humans have built by hand. They can literally test things to an extent that no human ever would.