Hello, how are you? Welcome to Exponencial, the podcast we make at Televisión Nacional together with Cenia, the National Center for Artificial Intelligence, where we open up the public conversation about artificial intelligence. We take hold of this fascinating topic. We want to greet Santander Open Academy, which accompanies us in every chapter and which is Santander Bank's global, online, free platform for learning and professional development. It supports education, employability, and entrepreneurship. For 30 years, with more than 1,000 online courses, programs, and scholarships, studying has never been so easy. Santander is the moment. Today we'll have a fascinating conversation, because in the pre-show it was like that, with two partners alongside a third. They are Chilean and they are the owners, the directors of the company Runway. I'm with Cristóbal Valenzuela and with Alejandro Matamala. Cristóbal is CEO and Alejandro is CIO, that is, head of innovation of the company Runway, created in 2018 in New York. Cristóbal, Alejandro, thank you so much for being with me.
Thank you very much for the invitation.
Yes, thank you for the invitation.
What is Runway? How did you meet, and how does a company that starts out doing machine learning models become a global leader in artificial intelligence for the world, mainly in audiovisual, film, and commercials?
Yes, I'll start. I'll start a bit with the story. It goes back almost a couple of years — 10 years ago we went to live in New York with Alejandro. We didn't know each other here in Chile; we met in New York. We went to study in New York for a master's degree, right? And the idea of the master's was mostly to study new technologies, and we got very much into what was then the beginning of this artificial intelligence revolution. Again, this is 10 years ago. Ten years ago, artificial intelligence is like 15 years in terms of... it's too fast, too fast. But the important thing is that we got in very early, very, very early, and we became interested not only in what was happening at that time, but in where it was going. And we started doing research, trying to use these models, this technology that was still very, very incipient, in industries that were still little explored at that time — for example, as you were saying, film, art, advertising, robotics. We started thinking about how these models could be used in that direction. And when we graduated from university, we started a company, the company that is called Runway, and we raised capital that allowed us to stay in the United States. And since then we've done a lot of cutting-edge research. We focus mainly on training and creating these models, and I can show you a bit more about how they're created, but we dedicate ourselves to training, creating, doing science, and then basically selling the models, delivering them to the companies that can use them. And we've been doing that for 8 years now. The company has been around for 8 years. We have offices in New York, in San Francisco, in Paris, in London, in Japan, in Seattle, and now we're looking at whether we want to expand to Latin America, which is why we're here.
Interesting. We'll talk about that later. How many people work at Runway, and what is the company's market value?
Today 250 people work in all areas, from research, business, product design — they're all distributed around the world, in the offices Cristóbal mentioned — and it has a valuation today of 5.3 billion dollars.
That's 5,300 million dollars.
How impressive, guys. You're in Chile because we're going to talk specifically about how it translates into films, for example, the films we see on Netflix, the films made by the big studios, the adoption of the software you develop, which is called Runway — I don't know if you call it software or the model or the robot — and that it's being used because there are also a lot of limits, right, union-wise, in labor matters, in ethics, in creativity, and so on. But you're in Chile with a mission, and for that you met with the government. What is it about, Alejandro?
Yes, today we are, as Cristóbal said, in a process of internationalization. We opened the office in London, we opened it in Paris because they are very strategic markets for us, just like Japan, because we already have investors and because there's a very important boom in what is physics and robotics, which is where we're betting now. And recently we were in Brazil, because Brazil is our fourth or fifth market that's growing very fast, and when we were there we said, 'Why don't we first go home and explore what we can do there?' Not necessarily or only with a commercial office, but perhaps with something more ambitious, bigger, where we could set up a research center and invite researchers to work with us, to develop advanced technology that we can deploy in different key industries of the country, such as mining, agriculture, aquaculture, which are sectors that will be crucial for the deployment and development of robotics in the future. We are working on new models that are capable of — that is, they are new technologies that use video models to teach robots to generalize different tasks in the real world. What does this mean? That we pass these videos to the robots so that they see, so that they see the world and learn how the world works and can learn to develop different tasks in different environments, different circumstances. So we believe that if we are able to capture that data here in the country and deploy it with robotics, we can start to create a productive and research industry in this area.
How much would it cost to build an advanced artificial intelligence laboratory in Chile, and what was the conversation you had with the government? Who did you meet with, roughly?
Yes. The truth is it's still hard to quantify. I think there are a couple of reference points. We already decided to invest close to 40 million dollars in Japan and in Asia. We're also investing a lot in France and in Europe. Depending a bit on the level of support we see in the region, we'll decide when and how we can, from our side, put in capital. Now, the most important thing — and in those countries did the state get involved? In some, yes. In England, not much in terms of hiring people, in terms of public visibility, and they have capitalized, the states have put money — well, they put money through incentives, through taxes, through... Exactly. Right. But clearly there are ways to... And there's another thing that's much more interesting, which we've also talked about with part of the government, which is that, for example, the government of Singapore is an investor in Runway. The government of Singapore decided to invest in Runway. Ah, look — very early, and this is a form of public investment that many of the most advanced countries in the world are taking. Poland does the same, as do Canada, France is doing the same, well, Japan, Korea, which are able to invest in companies or industries that are key for the country. And it's not a form of subsidy, not a form of co-investment, it's simply understanding that the best way to attract that capital is to invest in the companies. And so that's what we're also trying to see if it can be done in Chile, if other countries have done it, other countries that are at the vanguard — I believe it's necessary that Chile does it. Now, perhaps the most important thing in that whole conversation about why we're here, this internationalization, is important to recognize a bit — and this is what we've been studying for almost 10 years — that this technology is perhaps one of the most important technologies we've made as a society. And I'm not exaggerating about the level of economic, social, political impact, and the speed at which it's moving is quite, quite important. We have to understand that internet took 20, 30 years, 25 years to gain speed, to change how we function. I mean, now we're doing a podcast that's going to be totally distributed. Internet 30 years ago, you'd say impossible, right? Today it's normal. And that same speed of change isn't happening in 30, 40 years, it's happening in 12 months, and it's going to start... So, considering that the government and the public system especially moves much slower than the private system, a big part of what we're trying to do here when we meet with the government, with the minister, with the Ministry of Finance, with the Ministry of Science, Technology, of Defense, is a bit to make sure that the country is reacting at the speed it needs to react.
And they were left with that impression.
I think it's expert — I mean, we're having those conversations, but part of what we're trying to do is to make sure that from our position as Chileans in the center of the industry, we can bring that knowledge and make sure that our country is also thinking about what's coming.
Yes, how interesting. Later we'll talk about what practical applications it can have in areas like agriculture, defense, and so on. But let's go back to Hollywood, let's go back to audiovisual and how Runway has become a tool that they use — what percentage of the film industry, more or less?
Today I'd say that 95% of the biggest Hollywood studios are using us for conceptualization, production, or post-production of their audiovisual projects. Also the marketing and advertising industry — we work with the four biggest advertising companies in the world, where we're distributed in their smaller agencies that are everywhere. But today we see — we recently made a collaboration alliance with a studio called Lionsgate, which is a very big studio in Hollywood, in which we partnered to develop technology and develop models that were capable of producing creative outputs of quality for this industry. So we use their film catalog, their show catalog to train these models so that the output is very creative and very cinematic, but also to understand how this industry wants to use these technologies in their processes. We partner with their creative teams, with their production team to know at what moment they want to use it and in what way they want to use this tool.
Does Runway replace the job of a production manager, who has to do pre-production, of a screenwriter, of an editor or colorist? Because replacement is a very present issue in the world of creative industries. Now —
A bit of both, to be honest with you. Ah, the important thing is that we have to recognize that technology changes the way we do work, right? Technology changes the way we do tasks within your job. And I tell the story that when I worked here in Chile, I did my first professional internship, I had to do it at Banco Estado, and at Banco Estado, a building here on Moneda, very old-fashioned, commercial by profession. Yes. And this was in 2000, when I was about 18 years old. And I remember that in the elevator there was a man who pressed the elevator buttons. Yes. The elevator operator, do you remember? I don't know if... Right, to this day there are occasionally some, but before, all the buildings in the world and in Chile, of course, had that function. That was a job, it was a paid job, a job where someone was literally pressing buttons. Today obviously, of course, we don't need that job — today with sensors, and probably 99% of the elevators in the world are automatic, you don't need to have someone. That person, that person whose job was pressing the button, could, as a consequence of a better technology, which is the automatic elevator, move on to doing something that has much more personal value and for society too, right? Now that person could be studying and working in a job that is much more productive and much more interesting, right? Or they could be unemployed. That could also be, but we're thinking that ideally society moves toward new opportunities, right? I believe that substantially Chile is a much richer country than it was 80 years ago, 20 years ago, 10 years ago, when there was a market of people pressing buttons, right? A bit the same, the same thing is happening today, where there are many jobs and many tasks that are quite repetitive, manual, where probably an autonomous system, an artificial intelligence can help you substantially. In some cases it can do most of the work, and a bit of what we try to do when we talk with companies, when we talk with the worker conglomerates, is that we have to help not only to understand the technology, but to educate and move those people toward new tasks. A bit not only staying focused on pressing buttons, but understanding that probably that task is going to be fully automated. What do you do afterward? How do you educate yourself? How do you start to understand what new possibilities can exist? And what we've seen today is that a bit the opposite of what many people fear about artificial intelligence is that there will be few jobs. We've seen a bit the opposite, and in the United States this is documented, and there are data and statistical reports that say that companies that are adopting artificial intelligence faster than others are hiring faster. I mean, I tell you from our own company. We're hiring more people today than at any other moment in the company. Yes. Because we need people who can keep thinking about new things, about using technology — we can move much faster, making software today costs us the same as it cost before, cheaper, we can make more applications, we can sell more, start thinking a bit more that way.
But for the creative process, do you say that people are irreplaceable?
In some parts, yes, totally irreplaceable. We need... in some part of the process of the company —
Or in some industry, you say?
