Apple Podcasts Link
This episode features Thierry Klein, President of Bell Labs Solutions Research, discussing his journey from Luxembourg to MIT and Bell Labs. He highlights Bell Labs’ century of groundbreaking innovations—like the transistor, solar cell, Unix, and satellite communications—while exploring today’s frontiers: 6G, quantum computing, AI, digital twins, and space-based networks.
Thierry Klein: So I'm Thierry Klein. I'm president of Bell Labs Solutions Research. Bell Labs is the research part of Nokia, and I'm particularly focused on research into new technologies, new markets, new customer segments, and new business models that would expand the portfolio and the reach of Nokia.
Jim Barrood: Great. And so that sounds like a really important job. So tell us, let's go back in time, the way back machine and tell us how you got to where you are now.
Why don't we start with high school?
Thierry Klein: I grew up in Luxembourg, in Europe. Luxembourg is a really small country, but the size of Rhode Island. I grew up there, went to high school and was interested in math, physics but didn't have this big vision or ambition of what I wanted to do.
Just was just interested in math, physics, science, and chemistry. Then did my engineering, my undergrad, my master's in France. I have a dual masters in mechanical engineering and one in automation and robotics. And while I was doing my master's in France, I was on vacation on the East coast with my parents and traveled up and down from Boston to Washington and said, why don't I come to the US?
After I graduated in France and did a PhD in the US. I'd been to the US as a kid on exchange programs and so forth. So I always had a view towards us. But that trip gave me the idea that I should continue my studies and come to the US to do my PhD here as opposed to doing it in Europe.
So I came, went to MIT and spent a few years in Boston. Still rooting for all the Boston sports teams. Sorry. And then I actually switched fields when I did my PhD. So my PhD was in information theory, communication, and networking technologies. And when I graduated, of course, Bell Labs is the place to go for information communication technologies in the US and in the world.
So after I graduated I joined Bell Apps and here I am still at Bell Apps. Started as an individual contributor and. Over time I took more and more research management responsibilities and expanded the breadth of topics and technologies and business and strategy aspects that I get involved in.
Jim Barrood: Got it. Okay. Talk to us about, you mentioned individual contributor and that word has gotten a lot more sort of attention in recent years. Tell us what that meant to you back then and what it means these days in the world of tech.
Thierry Klein: I think when I joined Jim, it was, my first projects were really almost a continuation of my thesis.
It was the 3G days when I joined in early 2001. And at that point it was really just thinking how we optimized 3G networks. I did a lot of work on what is called scheduling algorithms, quality of service control. How do you handle multiple connections on the same wireless air interface? And really did research in that direction.
Worked with our business units to transfer some of the algorithms that came out of bell Labs research into the business units. And then slowly over time I ended up looking at broader networking topics versus just the wireless and interface. And I think an individual contributor on Bell Labs is always somebody who has the deep technical knowledge, the expertise.
And works on the project and, but never works really alone. Your individual contributor for us is also not somebody who has managing management responsibilities, but is really working on the technical problems, coming up with creative disruptive technologies, and then working in a team, whether it's another research team, maybe collaborating with people in university, collaborating with business units to transfer the ideas into a product path.
Jim Barrood: Got it. Okay. So tell us about Nokia Bell Labs. A hundred years innovating. It says right over your right shoulder an amazing history, right? It might take us hours to go through all of it, but for the folks who don't know much about the history, give us a brief overview of what Bell Labs has created over the years.
Thierry Klein: Yeah, absolutely. I think Bell Labs was created in 1925. The Bell Labs named Bell draws from Alexander Graham Bell, who invented the telephone, of course. And we've always been in the New York, New Jersey area from 1925 on now. It's a global research organization. We're part of Nokia right now, but historically we're part of at and t.
And then with the divestiture of Lucent out of it. Then Bell Labs research went with Lucent merged with Alcatel and then Nokia bought Alcatel Lucent. But through the mergers acquisitions in the telecom industry, it has always been the research part of the mother company. And I think what sets us apart is a few things.
It's both the breadth and depth of research we have. A lot of different disciplines that we're researching on, whether it's mathematics, algorithms, electrical engineering, computer science, physics, chemistry, material science, quantum computing. So, a lot of disciplines in our history and even today.
So, we have the breadth of technology, we have the depth of the technology, and we're always looking at solving the hard problems of our generation. Those problems change over time, of course, but we're looking really at what are the defining trends? What are the defining technologies of that generation?
What are the hard problems? We don't want to do incremental engineering, incremental research, but really thinking about the future. So, what should the world look like in 10 years? What are trends where the current technology path will not be sustainable? Maybe you hit lower physics, maybe you hit a resource constraint, whether it's energy or capacity on the wireless network or on the fiber.
So what are some of those big trends where you need inflection points on technology? And then we imagine what that should look like in the future, but then. Bring it back to today. What are the steps that we're taking to go to that solution in the future? So, it's not a baby step from here on, but you have an ambition of where you want to go.
And that may not necessarily be now, that may be 3, 5, 7, 10 years out. We talk more about the disruptive technologies that we want to create rather than long term research, because your long term might be my short term. And if you look at certain disciplines. Short term and long term are relative. I think about the wireless field between every generation.
It's about 10 years. So that's long term. That's the span of research in wireless communication. It's 10 years for one generation to the next. But if you're talking about ai, 10 years is a lifetime. So, you're really looking at maybe six months, 12 months, three years timeframe for having a disruptive impact.
So we're thinking it's very disruptive. We're thinking somewhat academically, but we're always grounded in an enterprise industrial context. So we don't want to solve problems just for the sake of it. We want to solve problems because they are meaningful for industry, society, and humankind. And then not just solve the problem, but demonstrate that the solutions have an impact in the real world.
Jim Barrood: Got it. And one of the most important inventions was the transistor. And I know there's a couple other home runs, but just talk to us about the transistor, how important that is. People just don't realize how fundamentally critical transistors are and how many there are in our phones and our blah blah, this or that, like TVs, et cetera.
Thierry Klein: Yeah. The transistor is still the single biggest invention coming outta Bell Labs. I think every year we're trying to do one better but the transistor is still the most transformative technology coming outta Bell Labs. And at that time in the 1940s before we had the transistor, everybody was using vacuum tubes for amplification of signals or for processing.
But vacuum tubes had multiple problems. They were large, bulky, unreliable, and very power hungry. So then there needed to be an inflection on looking at different materials, looking at semiconductor technologies to see how we really change how we do signal amplification processing and computation.
I think if you think about what I said about the Bell Labs mindset, the research was not to build a better vacuum tube. The research was not building a more power efficient or more reliable vacuum tube. It was just really recognized that the vacuum tube. Would not be the technology going forward, would not be on the sustainable path, even if you make it a little bit more reliable, a bit smaller and a little bit more energy efficient.
We needed a fundamentally different technology path, and that led to the invention of the transistor, which of course has completely changed all communication processing, computing it's in every phone. It's in every computer. It's in every electronic device. Just as a reference point, if you look at your smartphone today, it probably has something like 15 to 20 billion transistors in it.
And if you did that with vacuum tubes, your iPhone would be about the size of the Empire State Building. Just to give you a measure of you're not carrying an Empire State Building in your pocket, but it completely changed. What we can do, because we have these tiny devices that are now at an incredibly large scale in basically every device, every computer, every phone, every tv, anything that has electronics in it.
Jim Barrood: That's incredible. An incredible statistic. What about other, any other home runs or base hits that we might not be aware of?
Thierry Klein: Yeah, I'm not sure how much time we have Jim, but there's a long list of favorites. I think the telephone is, I always go back to the telephone because it's the first one.
It just fundamentally allowed us to communicate over long distances before you could only talk to each other if you were close. And the telephone allowed us to talk to each other over long distance. And I think that fundamentally changed how people. Interact and communicate with each other.
Talked about the transistor, the first practical solar cell was invented. Bell Labs, huge impact. Of course, the first communication satellite was invented at Bell Labs in 1962. We launched Tel One. Obviously not the first satellite, but it was the first communication satellite that was allowed.
Communication on voice data TV broadcast across the Atlantic, and again, it brought people from different continents together and they could communicate and in real time share news and information. If we go a little bit more on the. Software side. Bell Labs invented Unix CC plus programming languages, which are now the foundation of really every software and computing environment.
I think there's lots and lots of invention and for us the challenge is every day to, to come up with the next one. But there's a long list and I think a lot of these inventions. We're then also recognized with prestigious awards. We have 10 Nobel Prizes. We have five tour awards, which is the Nobel Prize of Computer Science.
And we have lots of medals from the National Academy of Engineering and Science. So lots of awards and everybody thinks about that and focus on that. But the point for us is that's never a goal. Nobody has a KPI to come up with something that leads to a Nobel Prize. Nobody aims to win a touring award, but those things happen if you do meaningful work.
If you solve meaningful problems and the research that you do leads to something that has a profound impact on society, these awards come naturally. And so we see it as a recognition for having done impactful work as opposed to a goal in itself.
Jim Barrood: So talk to us about today. What are you working on today at Bell Labs?
I think you, I think six G is one of those things. But tell us, let's focus on communications first and maybe computing.
Thierry Klein: Yeah. So our research is I think there's two parts to the Bellop DNA on the research. One is research in the direction of the portfolio of the company. So that's whatever portfolio of product solution service the company has today.
What is the next generation or two generations down the road of that same product. And so anything that's dealing with networking falls in that category. The company Nokia is selling, of course, 5G networks today for comms. service, private networks, enterprise networks. What comes after 5G? How do we increase the capacity?
How do we improve energy efficiency? How do we support more devices? So that leads us to six G. Actually, the research on six G is almost wrapping down at this point. We're still in the middle of the 5G cycle, but the research on six G has already started several years back and is now rounding down as we get into the standardization effort and so forth.
So that's really on, on the wireless side. Then we also have a lot of research on your fixed access, your broadband to your home technologies. How do we increase the speeds on that last mile to your house or to your business? Improve the speed of optical transmission and IP networking and optical networking, which is really the backbone of the internet and connecting all the data centers between the data centers and within the data centers.
And pretty much every year we beat our own record on the speed of an optical transmission system. And how many bits can you send on a strand of fiber? So it's really pushing that from a device perspective, power, energy, perspective, throughput capacity perspective in all aspects of the network technologies.
And more and more AI comes into that as well.
Jim Barrood: Before we focus on ai, what about quantum?
Thierry Klein: So quantum we see quantum computing as the next. Generation of computing. So maybe it's the quantum revolution, the way we invented the transistors, the transistor, a long time ago. Now it's the qubit and how quantum can really enable us to solve problems that are currently not within reach of classical computers.
So we have an approach called topology of quantum computing that we think is quite promising to build a utility grade quantum computer. It's not just about having a single. Qubit the quantum bit, but also making sure you have a large number of qubits and they're stable so that you can build a scalable computer to solve these meaningful problems.
So we put quite a bit of effort on quantum computing. We're also looking at quantum communication and quantum networking as additional disciplines. Quantum safe security, post quantum cryptography. So there's multiple. Multiple domains within Quantum.
Jim Barrood: And for those who aren't familiar with quantum computing, can you give us a sort of a basic overview or primer on what Quantum is and what it could mean to us?
Thierry Klein: Yeah. I think with Quantum if you compare it to a classical bit a classical bit is either zero one a qubit can be in in multiple states. So it's using the quantum. Mechanical properties of particles. And that allows us to fundamentally do more computation than we could do with a classical computer.
So anytime you have a problem that is combinatorial in nature, where looking at the solution space, it's a combinatorial, exponential number of options, a quantum computer will be fundamentally better and faster. At going through these combinations and finding the optimal solution that can be useful for drug discovery, for example.
It can be useful in logistics. It can be useful in energy grid optimization. Anytime when you have these large large computational problems that have a lot of. A, a large search space for the solution, a quantum computer will be more efficient and faster, and something that might take years to solve can then be solved in seconds or minutes.
Jim Barrood: And what's your best projection on how many years out will quantum computing be standard and upon us?
Thierry Klein: I don't think anybody in the near or medium term future will have a quantum computer in their desk or in their home. But having these utility scale quantum computers that can solve meaningful problems, I would project it to be between five and 10 years.
Okay.
Jim Barrood: Got it.
Thierry Klein: There's still a lot of work to be done on. Science and building that up. There are quantum computers today but they're susceptible to the environment and noise in the environment. And as a result, they need quite a bit of error correction techniques on top of the physical qubit.
And so we need to overcome that because you get into a problem of scalability and dimensionality if you have a large number of error corrections that you need to do to get to stable qubits. And we believe with topological quantum computing, we have a fundamentally more stable qubit. So, we don't need that error correction factor of maybe a factor a hundred or a thousand on top of it.
So it's all about building a quantum computer that has. Stable, effective qubits that can use for computation. And when we talk about stability. A lot of the qubits that are known today and that other companies and research institutions talk about and have shown they're stable on the scale of milliseconds, that's too short a timeframe to do meaningful computation.
We need something that's stable on the order of minutes, hours, days, and we believe with our approach on topological quantum computing, we can show that qubit would be stable for days and weeks. And so you don't need as many. But we still need to produce that and we need to produce device with multiple of these qubits, but we think it's a more scalable path and therefore a more practical path.
Jim Barrood: And so is it fair to say that quantum computing is the next big thing after ai as some people are suggesting?
Thierry Klein: I don't know. That's a big question to really predict what's the next big thing. I don't necessarily think of AI or quantum, right? AI is about understanding.
But you need a lot of computation. And then quantum is a new paradigm for computation. But I said communication, sensing, crypto has other aspects. So I think those are probably going to be the defining technologies of our times. But we should also not forget the networking aspect. Because if, the promise of AI is to be realized. You do need networking capabilities because AI is fundamentally driven by data. You need to collect that data. You need to bring that data together to crunch it through the AI models. So we don't think that the AI promise will be realized if you don't have the advanced networking capabilities.
because everybody talks more about the AI part. But AI at scale is not going to happen without advanced high-speed networking as well.
Jim Barrood: Got it. Oh man, that's really informative. So, let's get into ai. And you guys have been working in ai, obviously it's gotten very topical and there's been a lot of hype.
I like to say this is the fourth and final hype cycle since there's now consumer, penetration and usage, but tell us what you guys have been working on and what the impact that Bell Labs can have on the AI sort of environment.
Thierry Klein: Yeah. So, the first thing is AI for us did not start when chat GPT came out.
AI has been around, as you said, Jim, for a long time before that actually, Claude Shannon, who's the father of information theory, he worked on early AI systems and tools. Then Jan Koon when he was at Bell Labs, he he's the inventor of convolutional neural networks, which are really a basis for a lot of the AI machine learning today.
When he was at Bell Labs, he worked on a system for automatic handwriting recognition which was important at that time for the financial industry of how do you recognize numbers that you write? On your check and how do you recognize that automatically when everybody has different handwriting?
So, he worked on that. Now if we fast forward to 2025, what are we doing about AI right now? I would say our AI research has about four four main pillars. The first one is really just looking at what are the foundational technologies in ai? What are the new algorithms, the math, the models?
And really the technologies that, the components that drive AI going forward independently of the application and the use case. So really just the foundation of ai. The second one I would say is now we're looking at applying AI to networking and communication problems, which is very much core to the business of Nokia.
And there we see AI coming in through the entire lifecycle of the network, and whether that's on wireless or your broadband to your home or your IP and optical networking independently of the networking technologies or the networking domains. We see AI coming in everywhere and coming in through the lifecycle.
And when I say lifecycle, I mean it can be right at the design of the network. How do we make sure the next generation of networks are AI native? And that will come in six G for example. How do we not add AI later to the network? But how do we build AI right into the network design from the beginning?
So that's very much the design. Then you look at the deployment of networks. Where should we deploy base stations? How many, which capacity, which configuration? So, you get AI into the deployment of the network. Then after you deploy it, you bring AI into the operation of the network. When you think about a wireless network, the traffic is not uniform during the day.
At night, you have traffic maybe in different parts. You have less traffic during busy hour, rush hour. You have traffic in certain parts of your city, not in others. So how do you automatically adapt the network capacity to the traffic of the network? And one important aspect there is energy efficiency.
How do I use ai? To optimize the network for minimum energy consumption. So that's during the operational phase. And then maintenance, customer care, troubleshooting. We do some really amazing work on using AI for predictive maintenance on hardware. Can I predict hardware failures? And we've shown that we can predict hardware failures up to two weeks in advance.
Which of course has huge impacts. Now you can replace equipment before it actually fails, and then you reduce disruption to the service. So just how we see AI coming through the entire life cycle. So that's, let's say pillar number two. Pillar number three is pillar applying AI into industrial and enterprise applications, and how do we bring it into.
Physical environments, whether it's a factory, a harbor, a port, a hospital, university, campus, energy grids, how do we bring AI into solving business problems in those enterprise industrial environments, whether it's for productivity improvement, efficiency improvement, safety energy efficiency, and so forth.
So how do you understand that physical environment, whatever it may be, how do you extract useful knowledge and insights that allow you to optimize your physical operation, your physical environment, your productivity, your production environment and so forth. And that's an area that is really a growth area for Nokia to go into the enterprise and industrial sector, not just from a connectivity perspective, but also from these additional technologies to solve.
Business problems in those environments. And the fourth one I would say is AI has to be responsible, has to be transferable. Transparent has to be verifiable and replicable, especially when we apply it to these mission critical enterprise environments. We have to understand what it's doing and it has to be.
Build on this a solid, sustainable, ethical foundation. And again that's more transparent to the particular use case, but we really want to make sure that we're not doing AI for the sake of it, but we use the transformative power of AI, but we're applying it in a responsible way.
Jim Barrood: Got it. Wow. That's a lot. And you mentioned Young Koon before, he's now with Meta, right? Facebook, he's one of the top researchers. So clearly someone who is still one of the preeminent leaders and thinkers in this space, right? Yeah, absolutely. So that, I guess, is an important thing to realize that Bell Labs over the years has created an amazing alumni group that's doing amazing things, and that would take hours and hours to go through.
But it's really impressive. It's an
Thierry Klein: amazing alumni group of Bell apps in industry, in academia. And now with our centennial celebration it's also a way for us to reconnect with some of them and invite them back to our sites for some of our celebrations.
Jim Barrood: That's fantastic.
So the other area, which I know you're keenly interested in is space. So, talk to us about your efforts, initiatives that are going on right now. In the beyond space sector?
Thierry Klein: Yeah. It's definitely a favorite for me because it, for me, marries work and research with personal passion and hobby and interest.
What we're doing right now is really the focus of taking the technologies that you and I use every day on earth. That are in our daily lives, in our service provider network, in our enterprise networks, and take those same technologies and bring them to space. Because what has happened in space, we, I don't mean earth orbit necessarily only.
I mean going to the moon, going to Mars, going to deep space. And by and large, we don't have the high-performance advanced technologies that we use every day are not used in space and. The question is, why not? Why are we inventing new technologies, reinventing the wheel, using not all the investments that have been done in telecom that we're using every day?
Why are we starting from scratch or using legacy technologies, using technologies that have developed, been developed 10, 20, 30 years back? That is not the path forward, we believe, but we believe the path forward is to take the technologies. If you think about wireless, take 4G, take five. G. In the future, take six.
G. Take the cellular technologies that we, the billions of people and devices use every day. Take those same technologies, benefit from them and take them to space. Of course, you need to adapt them, you need to harden them for the space environment. But at the core, it should be the same technologies that we use every day.
And so that's what we've been doing for several years now. We've been on the journey with NASA and other partners in the space industry. To validate that thesis. And this is the year when very soon we will deploy the first cellular network on the moon and prove that we can take those same technologies and make the necessary adaptations and optimizations to deploy a cellular network on the moon.
And then ultimately, we believe that's the foundational technology that will be scaled up to support them. Human exploration on the moon, going eventually to Mars and will ultimately support robotic missions, crude missions and the lunar economy that might develop in the next 10, 15 years.
Jim Barrood: That is really exciting. Is and will, when, since you're, since you know a little bit about space, when do you think we will get to Mars?
Thierry Klein: I don't know if we'll accelerate going to Mars now or not. I think we'll get to Mars sooner than we think right now. But I also think we need to go to the moon first as a stepping stone as a learning platform.
We may need to go to the moon to prove out some of the tech, some of the technologies because it's still closer to earth. It's still. It's very risky, but it's still safer than being really in, on, on Mars. And then we may need to go to the moon and use the moon as a launching platform extract some resources that might be used for fuel, for example, to go to Mars.
I would think that within 10 years we'll be on, we could be on Mars if you think how long it took from. President Kennedy announced that we are going to the moon to actually be on the moon. It was seven years or so, seven, eight years or so. That's amazing. We have way more technology now than back then.
So if we commit to it, I think we could be on Mars within 10 years, but also believe that we need to go to the moon first and maybe relearn lessons that we learned in the sixties and seventies and unlearned. Then use that as a stepping stone and a launch platform to, no pun intended, to go to Mars.
Jim Barrood: Fascinating. We will see. Time will tell. But that's interesting you say that we should go back and unlearn our assumptions, right? That's an important sort of process.
Thierry Klein: I think we may have unlearned things. If you think about it, we went to the moon over 50 years ago, and we had way less technology than we do now.
So why is it so hard? Why is it so challenging for us right now to go back to the moon? When we have way more technology at our fingertips. We've done it before, so to me it feels like maybe we've unlearned some lessons or the mindset is different, the commitment is different. But it's, it should absolutely be within our reach because we've done it before.
Jim Barrood: For sure. All right. Let's talk about general trends or. What you're seeing in technology going forward. Anything that you wanna talk about that you folks have been researching or, what should we be optimistic about and concerned about?
Thierry Klein: So we touched on quantum already, we touched on AI already.
So let me talk about another trend that I am quite excited about, and that's something that. Would broadly fall on the concept of digital twins and digital twins have been around for a long time. But I think we're more and more going to see dynamic real-time digital twins of large complex systems.
And what I mean by that is take an example of a digital twin of an airplane engine. We've done that before. We have a CAD drawing of the engine of an airplane as a digital twin in some sense. Then we did this where you see the engine, you can rotate it around, you can take parts off and so forth.
That's not very interesting and doesn't really allow you to optimize your aviation operations, your airport operations, your airline operations. So I think we'll see larger scale digital twins. That is to understand that capture first and understand in real time what is going on. So we're not talking about the engine of a plane, we're talking about the plane itself.
We're talking about the airport itself. We're talking about the entire city. We're talking about the entire airline operations in real time. I want to know everything that's going on. I want to be able to zoom in and out at different levels of granularity. I don't need to know everything about every part of every plane at every moment in time.
Maybe when everything is going well, I want to know the overall operational performance of my airport. Then if there's an incident or there's a traffic jam that maybe I need to zoom in and I need to understand more of the details of what's going on. But I want to collect all this information in real time, like instantaneously.
Understand exactly what's going on. Extract that useful insight and knowledge that helps the decision making of how I improve the process. How do I get around roadblocks? How do I get around inefficiencies in my operation? How do I optimize my production line? How do I optimize the flow in a of goods, in a supply chain to what?
So I really think we should get these large scale in real time. Models of what's happening in the physical world, we move that model into the digital world. We use AI to extract knowledge and insights. So we're still in the digital world. We can then play what if scenarios. We can play out multiple options of how we want to improve things before we put it into the physical world.
So we can play these scenarios, we can do these optimizations in the digital world, and then when we know what we want to do, we put that back into. The physical world and we optimize and change the physical world. So there will be this loop between the physical world, the digital world, and the back to the physical world.
And all of that requires a lot of sensing and sensing technologies to capture the physical world. A lot of AI to understand it and extract the useful information. Not everything is useful for you at every point in time. What's the salient information you need? You as a person make the decisions of how you improve or a machine makes the decisions of how to optimize your process.
And then you need to feed this back. So you need sensing, tracking, monitoring capabilities. You need AI to understand what's going on. And then you need a lot of computing to understand and evaluate your different scenarios. And then you need actuators to put the decisions back into place.
And you see how. Networking, communication computing come together as well as with robotics and intelligent devices because your robots will be your eyes and ears in the physical world. They help you capture this information, and they will also be your arms and legs to implement actions and decisions in the physical world.
So I think there is a really interesting play between all of these and none of these pieces are new. None of these pieces have really worked together at scale in real time.
Jim Barrood: That, I'm glad you explained that, because I think you sketch it out really well. But for those people who don't know digital twins at all, give us a really super simple example of how they can be implemented.
Thierry Klein: I think it's a multi-layer picture of the physical world. It's a digital copy of your physical world, but it's not just a photo. It has semantic meaning and understanding of what's in the photo or in your video. But you can look at it from any angle. So if I take a picture of a conference room.
It's not just a picture, but it knows that there are chairs, there are tables, there are doors. It will understand what the physical properties are. It will know that the table is heavy and you need maybe two robots to lift it and move it. The chair is much lighter, so you can have one robot lifted, robot cannot go through a wall, but it recognized that there is a part of the wall that has a handle and a doorknob that's a door, and the robot could open that and go through it.
So it has this. Physical, this representation of the physical world with meaning and understanding. So it can if you want to optimize the conference room and you want to move chairs and tables around, maybe you need two robots if you want to move the table, but you need one robot if you want to move the chair.
So it has really this deep understanding of the physical world properties as well as not just. A picture of it or a movie of it, but really the meaning. And then you can reason. If you're saying, I want to have more people in the conference room I can play scenarios. Do I move the chairs?
Do I move the table? Do I move the table to the side? Do I take it completely out? So you can play these scenarios based on what your business objective really is.
Jim Barrood: I got it. Alright. This has been really fascinating, Terry. Thanks a lot for talking with us here today. We usually do just one thing, some short clips and short insights.
So let me ask you, what's one thing that everyone can take advantage of when they're thinking of AI tools?
Thierry Klein: I think the main thing with ai it's still very early in the journey on ai. As much as we've used and we've known about AI for a long time, I think putting AI to everyday use, it's still very early in the journey.
And I would say just be curious. Be curious with your guards up. See how AI could be used. Don't assume it's the hammer for every nail, but be curious. Try it out and just be. Be cautious about using it. But give it a shot and see if it works and if it provides the benefit that you expect.
I think there's a lot of potential, a lot of transformative power in AI, but at the same time, it still has to make sense. Doesn't have to be the answer for every problem but just be curious and experiment with it and try it. But make sure you have some responsible guardrails in place.
Jim Barrood: That sounds good.
That's really a great tip. What about one tip for entrepreneurs or folks in different industries who want to partner with Bell Labs?
Thierry Klein: Yeah. Reach out to us. Collaboration is really important for us. We have traditionally collaborated, of course, with the customers of the company.
We've traditionally collaborated with. Academia, but more and more we collaborate with industrial partners, startups. We're trying to be more entrepreneurial in our own approach to value creation and how we transfer the research assets out of the labs. Not necessarily just going to the Nokia business unit to commercialize the research, but also look at.
Partnerships, joint ventures, spin outs, licensing corporate venture approaches. So we are really trying to accelerate the cycle of going from the lab to a commercial scalable deployment. We can, we're looking to partner with anybody to help us understand business problems, to help us accelerate that innovation lifecycle and just learn more, become more entrepreneurial.
And I think the starting point is always just reach out. Don't assume we're not available. We were very much interested in having more collaborations and particularly in the areas and the locations. Where we are in New Jersey is one of our main locations. It's headquarters for Bell Labs.
And we just announced a new Bell Labs Venture Studio. With the New Jersey EDA. So that's also a great opportunity for us to transfer the research into new companies that might come out of Bell Labs. So reach out, don't be afraid, and let's just have conversations of how we can work together.
Jim Barrood: Great. Great. Love that. Okay. Normally, Terry, we end with a poem or a quote or a saying that's important to you. So what do you have for us?
Thierry Klein: I don't have a poem. Jim and I, maybe in my spare time I need to take a poetry class and next time I'll have a poem. I have a, and maybe a few quotes, but given the conversation we had in the topics we talked about, I would go with one of my favorite quotes, which is, if you want to predict the future the best or the best way to predict the future is to create it. It's also interesting because it's not really clear who said it for the first time. Some people even think Abraham Lincoln said it, although there's no historical trace that he did, most people attribute it to Peter Drucker.
But independently of who said it the first time. I think it embodies a little bit of a developed mindset and maybe the researcher mindset at large. We imagine a future that would be better. We imagine a future that solves the problems that we see. And sometimes it's even things that annoy us in our daily lives and saying, oh, I wish I had this, or I wish this works better.
But then rather than imagine it just create it, do it, and make it happen.
Jim Barrood: I love that. That's a great way to end and you guys have made so much happen at Bell Labs. It's really a remarkable organization and we're just lucky to have it in our country, in our, in the state of New Jersey.
Thanks again for your leadership, Thierry. Thanks for joining us.
