Welcome back onto theCUBE here at the San Jose Convention for the Gala Award show. This is the GSA celebration. It's the annual award for it. It's the Oscars of the semiconductor industry. Pat Gelsinger, CUBE alumni is here. Always has his finger on the pulse. Your DNA, Pat, goes way back. Welcome back to theCUBE. This is not red carpet. There's no red carpet, but it's kind of a red carpet.
Well, hey, I heard you were going to be here tonight and I just had to come.
Word got out that theCUBE's here and you had to just take your private jet in and come right in. Tell me what's going on because we've seen so much action in semis. And again, I think I've interviewed back 20 years ago when I first did a podcast with you at Intel, and then now so much has changed. What are you working on right now? I saw your interview on Yahoo Finance. You got some cool things happening.
Yeah. Yeah. Well, I think semiconductors are hot again. And clearly AI, the whole surge, but it's really all of the new materials that are occurring. We have cryogenics and quantum computing and so much going on, which fundamentally it all runs in semiconductors and some sort. And this week we announced xLight and xLight to next generation light source. And if you think about a fab, what's in the middle of it, lithography. What's in the middle of lithography? Light. And God said, 'Let there be light.'
Yeah. And the physics, the speed of light, we'll see that. The photonics is a hot area. We're seeing all these new interconnects. Take us through in your vision, what needs to happen for semis to kind of continue the thundering pace it needs to go to give the connectivity, to get the horsepower?
Well, clearly what we've done this week with xLight to me is an important piece because when you think about Moore's law, it was always three things, John. It was always more transistors, better transistors and cheaper transistors. But as we went through the EUV transition, the transistors are no longer cheaper. We lost the economics of Moore's law. So what we're doing with xLight is being able to not just be 13.5 nanometers where EUV is today, but be able to go to two nanometers or below in wavelength of light, really spectacular, but also has improved the economics. Higher power light allows us to make the wafers faster, which allows us to improve the economics, get the economics back into Moore's law. So this has been an important week, but also we see that, hey, 3D packaging is more important. New material types get to be more important. We need to move from copper to optical for all interconnect as well. So it really is a time, as we would say, we're rebuilding the entire technology stack. There's never been a better time in tech.
And the demand is massive. Okay. It feels like the PC revolution where, okay, the 286, 386, that generation is now happening at larger, larger scale. And if you factor in geopolitical pressure, the incentives... Take us through what the new normal is because most of the lay people, even now in the mainstream, AI infrastructure is gone mainstream. I mean, people are talking about chips and GPUs and XPUs.
I've been doing that... I talk about this for like 20, 30 years. So explain the light piece and why that's important and what that will be. What does that provide the value to? Explain to the end user who's like figuring out, 'I got an iPhone, I'm going to have a body device. I'm going to have a hyper-converged edge soon with wireless spectrum, unlicensed coming together.' You're going to have smarter devices at the edge. What's all this mean for... How does that get better?
Yeah. Well, fundamentally, if we go back to the basics of Moore's Law, put more transistors on a chip, make those chips smaller, lower the power consumption of the chips. And that's why, hey, in your pocket, you now carry what used to take a whole data center to do. It really is pretty spectacular. But can we keep making things smaller? Can we keep lowering their power? Can we keep improving their economics? And to do that, lithography has played a central role for the 50 years of the semiconductor industry. So how do we make it better? We have to attack the hardest problem, light. And making smaller wave light allows us to make the wires smaller, the transistors smaller, and improve the cost of it.
And if you were back in school as undergraduate, maybe imagine our PhD program, knowing what you know now, what you're working on, what science is important now to study?
Well, I think all of the hard sciences continue... They're back in vogue, right? We sort of had a decade where, hey, if you weren't a software CS guy, right? Well, now all of a sudden everything is going to enable us to solve problems in material science, biology, chemistry, physics, all of those become back to the center of what we can do for the future. So it really is to me, it's back to the basics, baby.
Yeah. Back to the basics. Physics, material science, engineering.
And I call it the trinity of computing, right? Where we're going to have classical computing, like we've been building and CPUs. We're going to have AI, which is that scale processing of AI models, large language models, but we're also then going to have the third leg, quantum computing. Where there's so many algorithms that are not computable in digital space that are only going to be computable in quantum space. And as that materializes over the next couple of years, okay, almost everything that we can write in the equation for will now become computable.
The Holy Trinity of computing laid out by Pat Gelsinger.
