“If the threat model you have in the very near term, next couple of years, is a social engineering threat model, then yes, the playbook we've been building out on know-your-customer, on monitoring the use of different services, is going to be extremely relevant. Eventually KYC will become KYA, and it's sort of know your agent.”

I've come close a time or

two. Or would it be at the level of synthesis? And this is maybe a good place to tell us more about some of the work you've been very directly involved in on the synthesis controls.

You know, imagine swarm of agents. What are they going to run into as they pursue this nefarious goal?

Yeah. Okay. So we're imagining that we've failed at the level of constraining the model with classifiers or keeping certain training data out of its hands at the training stage. And we're now at the point where you're trying to take at digital design. You've successfully designed some nasty virus…

So tell me more. For one thing, I was interested to see that you had, in the process of zeroing in on a couple of these choke points, evaluated like hundreds of different possible choke points. So I'm interested in what makes these so natural. I'd be interested in other things you learned from…