00:00Concerns about AI have been going on for a really long time. It really kind of accelerated when you had
00:07an Anthropic employee named Jacob Toxson come out and post a multi-part thread on X, basically saying that he
00:15didn't think either company was sufficiently safe enough or responsible enough to develop this powerful technology.
00:23But they weren't taking safety to heart, you know, and especially, you know, he's speaking coming out of Anthropic. His
00:32diagnosis was that Anthropic in particular feels like they are aware of these issues, but they are in a circumstance
00:40where everybody else is racing and they feel like they also need to be at the forefront.
00:44The fact that Sam Altman, the CEO of OpenAI, felt comfortable enough to promote Dario's essay and say, yeah, we
00:55would also, you know, I'm in support of this. I'm in support of a lot of the proposals laid out
01:01here. I mean, that's unusual.
01:02These are two guys that don't trust each other at all and they don't really like each other. But the
01:09fact that they came together, at least in public for this moment, means both CEOs recognize that the noise around
01:17safety and the questions and the concerns about transparency and oversight have gotten so loud that they can't ignore it.
01:25It is hard to know what exactly slowdown means. Is it, you know, that they're promising not to develop future
01:33technologies? Or is it that they won't release that technology to the world? What is the timeframe? What is safe
01:40enough even look like? And perhaps more importantly, what kind of resources does safety and security get from here on
01:51out?
01:52You know, I'm like, will the companies actually take safety measures that they are saying they will? Or will they
02:00lean back and fall back into the habits they've had today, which is to sort of privilege things like capabilities
02:08research? And I think that's going to be, for me, the key question going forward.
Comments