00:00 Speaker a
Meta is all incorporated into artificial intelligence. CEO Mark Zuckerberg said the company plans to invest hundreds of millions of dollars to build AI infrastructure. Do you have any spending on super intelligence payoffs? Hi-tech editor Dan Howley is now joining us. Dan.
00:24 Dan Howley
Yeah, Josh, um, these are some big, big moves that Meta is taking on. Two data centers they discussed, one was Prometheus. This is a 1 Gigawatt data center. Uh, other uh, it's over 5 uh gigawatts. For example, gigawatts are 1 billion watts. Well, the estimate is that you can run up to 800,000 homes in a year. Therefore, the amount of force these are pulling is absolutely incredible. And part of the effort here is building a super intelligence lab for meta. And the idea is to develop AI that transcends human intelligence. Well, they obviously use these to train their own ais, their llama ais. And it is being used in their advertising efforts and their recommendation engine UH efforts. So, this is all part of a kind of long-term meta strategy as you move on to AI. And it is worth pointing out that in addition to these kinds of capabilities, AI is also used as part of the push into augmented reality space of meta. Therefore, we cover all products. But this buildout is absolutely incredible. So one size in a lab will be when it takes up a large portion of Manhattan and is overlaid on it. So, to give you an example, it's the size of the second UH UH data center they're working on. And how big it gets. So it would be interesting to see what they have to offer, the impact of these, when these are fully constructed, become clear. Um, they are huge, resource intensive UH data centers. Um, you know, we need to see where this kind of lead is in the meta.
03:25 Speaker a
You know, Dan, obviously, as you know, there are a lot of fans on the streets for this, right? And those bulls often, Dan, they will argue that this is a company where AI investment is rewarded. And they point out that the user's time and advertisers are spending, and I'm interested in what you're making out of that argument.
04:06 Dan Howley
Yeah, I look, that makes sense, right? Regarding meta, it specifically points out how advertising customers use AI UH offerings when it comes to building small portions of their ad campaigns. And that's really the backbone of meta. It's a small advertiser, not you know, it's not a large company, it's a small business that makes up a large portion of its advertising revenue. And we've got that kind of technology and features there and they can put their products out there, and they've finally helped them and helped them so far. And as far as you know, you pointed out that users' time will be longer, meaning they see that it's a recommendation that they helped a kind of engine that helped them raise more people. This looks like a poster child in a place where AI investments have come to pay off. Well, you know, we also see it on Google, Amazon, and Microsoft. They're more concentrated, but Microsoft is on the enterprise side of things. They have an ad. But when it comes to meta, it's really advertising.
