Soon $2 Trillion spent.. who is capturing the value of AI right now?

When you go back at the prior transformations, whether that was the, you know, transition to mobile or the transition to cloud, again, it always comes in waves. The first few, the first phase and wave, is always at that kind of infrastructure layer, because you have to build the foundations first. People have to spend upfront to be able to then go and build cool products. So, most of the returns in the first phase tend to be at that infrastructure layer. Yeah. Yeah. Then you move to phase two where it's cool, little applications, mobile apps, software, AI, and that's when, that's when the next layer comes and the and the returns start to shift in terms of the proportion of the return, starts to shift towards the, you know, the next layer down. Um. And I just think again, for this to be true and for this to not be a bubble, which again, I don't get me wrong. I I have absolutely, you know, I am one of the most convicted people you will find that it's not a bubble, so let's just make that clear and obvious right now. Let's have it right. Um. But I do think that, ultimately, it we it is going to need at some point to extend beyond the hyperscalers who are getting all the returns of AI, because when you look at it so far, we talked about the compute layer. When you think around the kind of product to model layer, most of it for public companies will come to, you know, OpenAI and will come to, I'm sure it's clear, obviously, but for public companies, most of it has gone to your Googles and in your Geminis. It's gone to, um, even Copilot seems to have driven