 We are in Albany, New York, at the most advanced collaborative semiconductor research facility in the world. This kind of facility is the products of billions of dollars of investment and over 20 years of dedicated work. It has been built on a highly successful public-private model designed to support an ecosystem of world-leading semiconductor suppliers and manufacturers. It is where IBM researchers have done some globally-altering scientific and technological breakthroughs in nanotech and semiconductor research. Many of you may not know this, but my early research at IBM was in the field of nanotechnology and lithography. You probably have heard about the field of nanotechnology. The field of lithography is a bit more obscure, but it is at the heart of why computing gets better and cheaper decade after decade. Mukesh Kare, one of the most respected leaders in the field of semiconductor research, is going to share how we're building the chips of the future. We're going to get a chance to talk about one of my favorite subjects for hard tech, semiconductors. Maybe we could start by discussing how are chips made. Well, Dario, it may not be immediately obvious, but transistors, the basic unit of computation on a wafer, is printed. Wafers are sawed out of cylinders of pure crystalline silicon, then polished to mirror-like finish. With lithography, we are able to print and ultimately etch billions of transistors onto a wafer with atomic precision. A high-energy laser fires on a microscopic droplet of molten tin and turns it into plasma, emitting extreme ultraviolet light, which then is focused into a beam. We reflect beam of a masked pattern that contains the complex design of the circuitry of the chip we want to print. That pattern of light is then shrunk through an array of atomically precise reflective mirrors, finally casting onto the silicon wafer at a microscopic level. This light exposure burns the pattern into the photoresist and after it is developed, it forms a relief pattern that can be used to etch the desired structures into the silicon. The wafer gets processed subsequently and cleaned to remove the resist. This process gets repeated layer after layer, as many as a hundred times. And over days and weeks, we get to ultimately create a fully functioning chip with transistor dimensions at a nanometer level. Imagine we're stepping down from our typical field of view by powers of 10. At 10 centimeters, our field of view crops the edges of a device like our phone. We see a printed circuit board as we move down to 1 centimeter, entering the chip and its 50 billion transistors. At a millimeter, we're now roughly to the point at which the industry had scaled 20 years ago, 30 million transistors on a single chip. Stepping down now to 100 microns, we see the largest elements of the integrated circuit. Just beyond the scale in our biology would be at the width of a single strand of human hair. Now at 10 microns, you'll see a portion of the chip with a large array of devices performing core functions of the chip. Passing beyond the nucleus of our cells at 1 micron, we continue to travel down deeper till reaching 10 nanometers, the scale of the very fabric of our make-up. So, Mukesh, there's a lot going on in this fab. Tell me about it. We use lithography technology to build the basic device for computation called transistors. Working together with our talented research team and many partners, we have pioneered a new transistor structure. We at IBM call this nano-sheet, which has become the foundation for every chip manufacturer's future chip generation. The nano-sheet structure is formed by vertically stacking multiple layers of silicon sheet channels around 5 nanometers in thickness, which is about 2 DNA molecules. In 2015, we were the first to create the world's first 7 nanometer test chip. A few years later, in 2017, we did it again, creating the 5 nanometer test chip where we first introduced the nano-sheet technology to the world. Now, I'm very proud and excited to say that we've done it again, creating the world's first 2 nanometer node chip. And we did it right here in this facility. There are almost 10 times more transistors on this wafer than the number of trees in the entire world. Scaling to this 2 nanometer framework will equate to 45% performance improvement over today's 7 nanometer chips using the same amount of power or a 75% power savings at the same performance level. Extraordinary, Mukesh. It's hard to imagine that this technology could get any better, but I'm sure you guys are going to try. Yes, and we like the challenge. These continued technology advances ensure an enduring platform for both our own hardware and systems, but also the entire technology ecosystem. While the commercial availability of 2 nanometer processors is still several years away, the IBM Research Innovation Pipeline gets directly commercialized through our hardware platforms. In fact, IBM's first commercialized 7 nanometer processor, based on our 2015 innovation, will appear later this year in IBM Power 10 based systems. And next, we will show you how these advancements are bolstering the capabilities of our system Z. Our work here underscores the importance of advancing semiconductor chip design and performance across all modern computing architectures. And investing in these innovations is also critical for our partners, such as Intel and Samsung. It is also vital to the secure chip supply chains of industry, from IT to car makers and to the success and security of our nations. Our 2 nanometer breakthrough will create advanced nodes that give hardware designers a more powerful canvas to create specialized tech. 2 nanometers is now the foundation for researchers to explore the future of hardware, including AI hardware, that can drive greater performance across everything we do. But we face a challenge. Today's AI is incredibly power hungry. AI's rapid progress has fueled an insatiable demand for computing power for ever larger neural network models on ever-growing datasets. We have to figure out new methods for running these large AI models on today's most advanced machines sufficiently. In other words, it requires hard tech. IBM Fellow Donna Dillingberger has been innovating in systems for years, and it's here to tell us more about what makes system Z such an amazing machine. Hey Donna, how are you? Good to see you. Hi Dariel. So you're a world-class systems researcher, so tell us what makes IBM Z this system so special. Z is known for its speed and its scalability. The Z15 was built with 9.4 billion transistors. With that type of power, we can run open-shift workloads at 4.7x higher performance, 4.4x lower latency with 34% less cost. It also runs 1 trillion transactions a day. It does that with half the energy that other servers require. It's the greenest server in the planet. It also has the highest availability. Z stands for zero downtime. There are clients that have never had an unplanned outage in years. While a Z server is running, you could pull out its memory, its processors, its IO drawers, and it will still run. Yeah, what a marvel of a machine in terms of transaction processing. But what about the world of AI? What can it do for AI? AI could be done in Z with millisecond response time. You have these transactions just being thrown at the server, 20,000, 30,000 transactions a second. No other server could do that and run AI embedded in real time. Wow, Dana, the green transactional power of Z combined with AI and security makes it like a marvel of engineering. So I want to thank you, Dana, for the amazing work that you do and that the team does to make the best computer systems in the world. Thank you and see you soon. Thanks, Ariel. The progress being made to advance the performance of classical computing is truly amazing. The two nanometer node chip we explore represents the absolute cutting edge of computing technology and proves that the power of bits continues to be remarkable. The way we are pushing computing to excel in AI workloads will allow pervasive and interconnected intelligent systems to provide extraordinary value. And these advancements are not just science for science sake, but for real business outcomes. It's how our research provides the technological innovation behind IBM's most advanced systems available today.