Justin Drake

two flavors of cryptography, if you will.

There's real-time cryptography, where you're just signing messages in real time, and there's no material impact on the actions that you made in the past.

And I think here, upgrading to post-content cryptography should be relatively straightforward for most of the internet.

There are some exceptions.

For example, if you have new satellites that have already been deployed and you literally can't upgrade them, then they will be signing

producing signatures that can be forged, but that's more of the exception.

Then there is another problem which has to do with encryption, where if there's material that has been encrypted today and you're not using post-quantum secure encryption today, that means that this data can be decrypted in the future.

And there's this whole class of attack called harvest now and decrypt later.

I think it's realistic that we're going to have

mass decryptions in society.

So we might have lots of signal messages from several years ago, or maybe lots of telegram messages or whatever.

I don't want to pick on one specific platform or maybe like troves of Gmail messages or all being decrypted simultaneously.

And I think that could have a very significant impact on society.

Yeah, so on the cryptography, I'm fairly confident about the soundness.

And the reason is that you can prove mathematically that your cryptography is correct.

So cryptography is a sub-branch of mathematics.

And there is this one exception where you have these hard problems.

And what you try and do, generally speaking, is that you calibrate, you parametrize these hard problems so that if someone were to computationally break the hard problem, it would use more energy than there is in the solar system or something ridiculous like this.

And going back to the cryptographic foundations that we're suggesting for post-quantum Ethereum, which is hashes, it doesn't get any stronger than that.

And this is, in some sense, the weakest cryptography that you could hope to have.