 IBM Edge, this is SiliconANGLE's exclusive coverage of IBM Edge. This is theCUBE, our flagship program. We go out to the advanced extract of Sylvain from the noise. We'll talk to CEOs, we'll talk to VP, senior VP's customers, CIOs, research fellows, anyone who has a signal and will extract that from the noise and share that with you. I'm John Furrier, the founder of SiliconANGLE. I'm joined by my co-host. I'm Dave Vellante at wikibond.org. Bernie Meyerson is here. He is the vice president of innovation and an IBM fellow. Bernie, welcome to theCUBE. What makes IBM different? Talk about that a little bit. Well, if you go back to the beginnings of IBM now over 100 years ago, it's interesting to keep in mind, we used to make cheese slicers, weight scales, meat grinders, really good meat grinders, I might add. If you think about how far you've come from making devices that make sausage to a company that builds the IT that took a man to the moon, or how far we've come from taking somebody to the moon by the way, 2K of memory, right, 2,000. Those of you who don't know, that's thousands, right? As opposed to 64 gigabits of stuff that you guys, you know, browse if you don't have. I mean, think about the dramatic innovations that had to occur to go from thousands to billions to go from cheese slicers to systems that essentially, you know, run the world financial network. I mean, it's a tremendous amount of innovation and it's continuous. And there's a lot more coming. What's happening now? What is the intersection of the computing side data and the material science side? Is there... Yeah, Moore's Law basically has died quite a time ago, actually. Gordon, who's a genius, has got this worked out. I mean, Gordon, of course, understood, if you put twice as much on a chip every 18 months, things are great. But what happens, of course, is nobody, including Intel, actually believes it'll continue because you get to atoms and they don't scale. So that's nonsense, of course. What will happen, though, is things will continue to improve, but they're gonna go down different roads. And that's where it gets very exciting. If you think about it, for instance, silicon goes fully quantum mechanical at about seven nanometers, which is a fancy way of saying it doesn't work worse spit. So, once you get there, and that, by the way, if you keep going at the present rate and pace, you're talking only seven to 10 years, which is nothing in the time scale of what we've done, you're gonna start looking at alternative materials. You're gonna start looking at things like carbon nanotubes and silicon quantum dots and all of the other materials, but in addition to that, there's a lot more to it than just shrinking silicon. I mean, IT has made progress, but 80, 90% compounded annually, of which a relatively small part is the silicon performance itself. So don't abandon hope, quite the opposite. You're gonna get tremendous integration. You'll see 3D where you're gonna be building what looks like a silicon cube. I mean, you're talking 10 layers, 100 layers, maybe 1,000 layers, where you have memory, you have communication, you have computing. There are enormous challenges to be overcome, but this will not stop. It will just go down a different road. Throwing money at just building fancier fabs isn't gonna do anything. And whether it's computer science or just science. What are the things that are exciting you? I think cognitive computing in the very beginning of it right now, I think is the most exciting thing out there. Cognitive started out where you were working on, for instance, just winning jeopardy. I think about this, you're taking on the two human champions at a Q and A that allows for jokes, and yet you beat them. You didn't just beat them, you're cream. Jokes, colloquialism. Exactly. Think about the complexity of that. Now extrapolate what you can do with that kind of learning capability because it's not just that it got it right, but if you get a wrong answer, it actually goes back in and reconfigures to essentially change the parameters that led it to the wrong answer and basically correct itself over time. Now apply that to medicine. Now apply that to finance. Now apply that to telemark. Apply it anywhere you want. And the quality and the interoperability with humans goes up dramatically. And that becomes fascinating because imagine a physician. A physician can only read four or five hours a month worth of papers, best case. Meanwhile, you have a machine. You have to turn the treadmill all day long reading. Exactly. And you would never touch one tenth of 1%. Whereas Watson can suck this stuff down at almost an unlimited rate. So imagine that you bring your kid in and it turns out that your kid has some god awful parasite and this doctor has never seen it. There are two papers in history written about it. You've got all the symptoms lined up, but what are the odds that physician is gonna find it in time to save your kid's life? These basic diseases that are sort of orphans that don't have coverage, they actually show up very, very quickly in a system like a Watson. And it will not only tell you how it got to guess what the disease is, it'll tell you what the tests are to make sure. It'll even say, if you do this test, it'll get me better odds. I mean, you can actually begin to work with the machine as an assistive. That's a dramatic change. What are some of the, what can you share? I mean, because you're talking about machine learning, going to cognitive learning machines, talking about different aspects of particles and chemistry and physics. Well, what I described is already there. So there's not a hurdle there, folks. That works. That's kind of neat. However, you go to the next level, you don't want to program a computer. Right now, if you want that to do something, you program it. I don't want to program things. I want to teach them. Okay, do you have any idea how useless programming is by comparison to teaching? Think how you learned. You went to class, you sat down, you watched something, and lo and behold, there it is. You got on the web, you looked at that. You understood. Speech to text, speech to C, speech to Python. And so why, you know, that's your point. And so the key here is imagine getting to the point that a machine actually has the same sort of IO that you have. Now that gets interesting. In other words, it is cognitive. It literally has the senses. It has sight, smell, touch, taste, hearing the whole bit. That gets really interesting because then, for instance, you're sitting with a patient and there's this assistive device next to you that goes, damn, I smell a staff infection. Or I smell, ah, this person has diabetes. It's not a joke. I mean, you can get there. How has collaboration been brought in to IBM's innovation model? Can you talk about that a little bit? We have a whole long, long list of ways we do that. We created communities basically around some of our sharing capabilities for Lotus and other things of that nature that enable you to very, very quickly find the expert. We have a technical community of 220,000 individuals. When you have 220,000 technical wizards, you know, they don't all know each other. You know, the difference, right, between the extrovert and the introverted geek, right? The introvert talks while looking at their shoes. The extrovert looks at your shoes. Yeah, that can be a problem. I'm not sure where I fit on that scale. I think it's part of the right. But the point I'm making is you really have to have ease of contact. But imagine if you create a community, a technical community, and then on top of that, you isolate, perhaps, let's say, the one top one half of 1%, which we have called the IBM Academy of Technology. It's sort of like our national academy. These guys actually self-select. They self-select. These are the top, literally, 1,000 out of 200-some-odd thousand people. They all form a tight-knit community, and we all know each other. And, you know, the joke at IBM is I'm sort of head geek. You call me up, I don't know the answer, but I promise you I know somebody who does. Young kids, high school level, okay? Or maybe even eighth grade going into high school. Just anecdotally, what's your advice to them? What should it be that you think about? Not what course to take, but like mindset. What's the mindset of this new generation? Just go do it. If you think about it too much, it ain't gonna happen. Any discipline advice, like math, science, the normal stuff? It depends what you're good at. I mean, there's no one answer, I'll be fine. Don't shy away from what you're good at because it may or may not be socially acceptable to work. Yeah, and the other thing is don't be afraid to ask for help. I've done a number of huge projects and I will tell you the most successful, actually the only ones, are when you just gather the best and brightest. You don't worry about this. If you do something that blows the world away, there is enough credit to last for everybody. Just go do it. You assemble whatever it takes to get the job done and then go ahead. I'll shoot. Okay, this is theCUBE, Bernie. Thanks for coming on and sharing your insight from education to work with the academy. It's great, and also the leadership at IBM, world-class and that always has been. Congratulations and can keep it going. This is theCUBE's SiliconANGLE's flagship program. We'll be right back with the CIO of Intel, Kim Stevenson, next to share with us the modern enterprise IT look and feel. We'll be right back after this short break.