Haseeb

You know, I don't know.

That's not, that's like, I think these, it's very clear that everybody who can have this wants this.

Sure, sure.

So are smart contracts fucked?

Some of them, yes.

Like the problem with many smart contracts is that they are immutable.

It is hard to change them or impossible to change them.

I think the first place that I would be thinking about is not the smart contracts themselves, but about the blockchains.

Smart contracts, in some sense, they are simpler.

They have less surface area, believe it or not.

Blockchains are enormous.

They are extremely complex distributed systems, and they almost certainly have bugs and a lot of surface area in all of them.

That's why I was somewhat disappointed to see that they weren't directly invited into Project Glasswing.

You know, the Bitcoin Core devs or the Ethereum devs, Solana even, I think it's pretty clear that these are probably some of the highest risk areas

projects because of the fact that they are they have so much surface area and it's so much harder to test them programmatically compared to smart contracts which can be formally verified in principle most contracts are not but the complexity of formally verifying an Ethereum client is just insanely larger than for any smart contract so I'd be thinking first and foremost about the blockchains themselves like how much more catastrophic would it be if there's a

you know, Ethereum inflation bug.

That would be massively more catastrophic than any single smart contract getting hacked.

So I would start there, but it's clear, you know, in the tweet that you referenced, I said, this thing is like COVID, but for software.

I think that's the way in which we are going to be thinking about this new generation of models in cybersecurity.

It's often the case that these labs create models that they don't release to the public.