Agent speed limits, such as tool calls per minute, should be implemented to prevent AI agents from operating beyond the pace humans can monitor, which is essential to avoid gradual disempowerment.

15:19revealed 10 years of algo research uh no training and then all training all at once. So another real um simple idea that I've been kicking around for a while is agent speed limits. And I thought this was a really interesting juxosition too over the last few days. is obviously with everything that's gone on at OpenAI, you would think um you know they wouldn't necessarily be rushing to raise by an order of

15:54magnitude the the pace at which their models work. Um and yet then we saw this like fast mode, ultra fast mode, whatever they called it, where I believe they said it was up to 14 times faster if you're willing to pay for that uh high-end speed. And you know, we're going to talk to Adam Wentchel from Arthur is going to be our first guest today. And I'm interested to

16:18hear his take on this speed question. But it does strike me as one of the biggest advantages that the AIs have relative to humans is just that they can work so much faster. And if we go another order of magnitude and just let them, you know, have tool calls flying around at, you know, the fastest pace that the chips can support, uh, if you're willing to pay up and, you know, sacrifice batch

16:45size and all that kind of stuff. It does seem like that's going to be really tough to monitor. Certainly, it's going to be beyond the pace at which humans can really make sense of it. So we're just like at every kind of decision point it seems like we're h hurtling down this path toward agents have to watch agents like what's the answer right I mean in the open AI

17:08thought it's going to be well we'll just have another agent you know watch this and then you know it can work at agent speed so that won't be a problem that one can also be 14 times faster perhaps but this is sort of the path to gradual disempowerment if nothing else where we're just spinning so many agent plates and it's, you know, it's agents watching other agents and we can't even watch the

17:33watchers because they're in turn moving so fast. I think agent speed limits is another thing that I'm I'm like uh pretty interested in developing as a concept. You know, I don't it's it's always tough to define these things, but tool calls per minute, I think, might be an interesting way to just try to make sure that these things are not like overwhelming systems. Uh, you know, moving so fast that the kind

18:03of async processes that are meant to keep track of them get left in the dust. Otherwise, it just seems like we're going to have, you know, more and more of these incidents popping up and they're going to happen at kind of flash speed and we're going to be like, boy, you know, that that agent called a thousand tools in 60 minutes and, you know, look at all that it accomplished

18:25and like nobody, you know, before we even kind of finished our first cup of coffee in the morning and, you know, kind of locked in for the day. It's like they can cover a lot of ground. Uh yeah, I think I think but that is both on the pro and on the con side, right? I think I think to some extent um if you're going to put them to work on