On this episode of Recode Decode, hosted by Kara Swisher, PARC CEO Tolga Kurtoglu dropped by the studio to talk about what’s new and what’s next at the storied Silicon Valley research and development company.
You can read some of the highlights from the interview at that link, or listen to it in the audio player below. We’ve also provided a lightly edited complete transcript of their conversation.
Kara Swisher: Recode Radio presents Recode Decode, coming to you from the Vox Media Podcast Network. Hi, I’m Kara Swisher, Executive Editor of Recode. You may know me as someone who invests heavily in R&D, which stands for retweets and dark sunglasses, but in my spare time, I talk tech. You’re listening to Recode Decode, a podcast about tech and media’s key players, big ideas, and how they’re changing the world we live in. You can find more episodes of Recode Decode anywhere you listen to podcasts, or on Apple Podcast, Google Play Music, TuneIn, Stitcher, SoundCloud, and more, or just visit Recode.net/podcast for more.
Today in the red chair is Dr. Tolga Kurtoglu, the CEO of Silicon Valley research and development company, PARC. Previously known as Xerox PARC, its team has pioneered some of the most important advances in computing over the past 50 years, including everything, including the UI, all kinds of things like that. Tolga became the CEO in January of this year. Before that, he oversaw PARC’s research into artificial intelligence, machine learning, security and more, all the key things that are popping up now. Tolga, welcome to Recode Decode.
Tolga Kurtoglu: Thank you, and thanks for having me.
No problem. So, let’s talk a little bit about your background, how you got to PARC. People like to know where people come from. Obviously, taking over PARC, which is the iconic font of innovation from the beginning of the Silicon Valley, is an important job. Why don’t we talk about how you got there?
Sure. I got to PARC in December of 2010. It’s almost been seven years. I think it will be mid-July is going to be my seventh mark. Background wise, I was born and raised in Turkey, came to US in 1990 just to pursue a graduate degree in mechanical engineering. That’s what my training is in.
No, in the U.S. I did my undergraduate in Turkey. I was born and raised there.
Always a nerd?
Yeah, pretty much. At the end of my undergrad, I decided to come to U.S. and get a graduate degree, so I did that in Pittsburgh, Pennsylvania at Carnegie Mellon. From there, I moved on to Austin, Texas. I lived there for a couple of years. I worked as an engineer, and then moved to the Bay Area in the summer of 2006. So, I’ve been in the Bay Area since then, and, like I said, I started at PARC 2010.
Why did you go to PARC? You were in Austin, I assume, working for either Dell or ...
Yeah, I did work for Dell. That’s a good guess. Number of reasons. So before PARC, I was at NASA Ames Research Center. I was a researcher there. So, I did combination of those, gave me a perspective on ... I knew academia, I have a lot of friends and colleagues that chose the academic career path, and I saw the government agency side and how research gets done in that context, as well as a pretty fast-paced product company in Dell, right? So ...
I think the synthesis of those experiences led me to think about my next career move, and PARC was a perfect choice for that.
And why was that? Because PARC has seen its day, right?
It has been there all along, but it really had its big time when Xerox owned it.
That’s true, but that is also what people remember PARC by. We do work on ... And I’d like to talk about that later in the podcast, about the kind of things we’re working on today. I think there’s a couple of reasons. The major one is it’s a unique R&D place, in the sense that it focuses on what I call translational research. Really kind of at that boundary of research and development. And you think about science and technology, but you also think about how to really translate that into market and marketing in a product and services, etc. So, it has that applied nature, and you can kind of see the impact of your ideas and your thoughts in your work. When it goes through that transformation translation, it creates that kind of impact that we want to create in the market in the world.
There is a big reward for a lot the people that we have at PARC today, if not all, but to me personally that was a big passion, right? So to really see through that ...
So you wanted to go there while you were working a Dell. You were doing what? What were you doing there?
Product design for what? For just computers?
For server products, basically. Yes.
Okay. And why did you want to move out of that to PARC?
Because ... I went from that back to the research realm, right? So I think there is a certain level of repetition, if you will, that it gets ... Okay, so you move from this product to the next product, but if you think about really the skill-set and the kind of problems that you’re working on, they get somewhat boring after a while.
And I always challenge myself to go out of my comfort zone and try to think about the next thing, and what might be the thing that would push me harder, compared to what I’ve done in the recent past. In that first transition into the research and development world first, and then go to PARC, because there’s a big variety of domains and problems that we work on there. And, as I said, that applied nature of it, perhaps a contrast to academia, where you solve the hard problems, and you publish, and you generate the knowledge, and that’s the mission of educational institutions, but then you move on to the next hard problem, right?
But I think there’s a big reward, at least for me, and a big personal passion, where you don’t just stop at the output of that R&D activity, if you will, but you really think about what that means for the people and for the society and for the culture, where it goes further down.
Sure, where it goes. You go to PARC to work on areas that are now the hottest areas in Silicon Valley. Talk about ... It was somewhat early to be looking at those.
Before PARC, again, I worked on a lot of predictive analytics and AI technologies, and at NASA as well, and a lot of my background and training is in engineering and how to apply, again, AI and other computational technologies to engineering problems at large, and with a specialization in product design and development, and how to do that better. What kind of tools and technologies you might develop to do that more efficiently, and to come up, both from a process point of view, more efficiently, but also to be able to design and develop and deploy better products.
Right. At the time of my transition from NASA to PARC, PARC had won a big contract from the Department of Defense, DARPA, agency that really looked at how to revolutionize how people go about designing these highly complex systems. That was a perfect fit for my background, so I made the switch. I started working ...
This I the Defense Department’s research arm?
Right. And so you went there to do that? So, they had contracted with PARC to do that?
And what were you trying to do with that? What was the concept, specifically?
Sure. I can talk about that. So the whole idea is when you develop these complex systems like automobile or airplane and whatnot, there is thousands and thousands of — tens of thousands, essentially — of requirements that you have to meet, NGA requirements, compliance, etc. And if you look at the typical design cycles from ideation of concept to coming up with initial designs and what we call validating, and verifying that, and then manufacturing and deploying, it can take decades.
And the aim of DARPA was to ... It can take decades for a number of reasons, but partly it was because you have to do a whole lot of testing before you deploy these systems. And a lot of those are what we call “hardware in the loop” testing where you have to build prototypes of these things and test things.
Like crash simulation, right? So you have to drive the car to the wall and then see what happens.
It crashes, and you need to assess that. But the beauty of the developments and computation and modeling and simulation allows us to do a lot of that in a virtual environment as opposed to a real physical environment, right?
So that was the idea of how to really leverage computers and computing power in a way that would do two things, that would allow you with new tools and methods and also workflows conceive and construct different alternatives for a given design, but also use virtual environments and simulation to be able to really validate and verify that it would meet your requirements to your satisfaction in a much, much shorter amount of time. So they really wanted to go from something that takes decades to something maybe that takes months.
Months, a shorter time to do the computation. So you go to PARC and you start working on these things. How big was PARC ... give us a sense of how big it is now, how many people are there. Where is the budget from, besides secret government agencies?
Good question. So we have a little over 200 people at PARC today. It’s a combination of our technical staff, which is ... the majority of them are PhDs coming from different disciplines. All the major engineering fields, aerospace, mechanical, electrical, industrial scientists, a lot of computer scientists, as you might imagine. And physicists, chemists, etc. But we also have social scientists, ethnographers, anthropologists, and user experience, user studies experts, as well as business experts that understand certain markets, IP strategy and those kinds of things. So we really blend those set of talents and resources when we ...
So kind of like university without students there?
We have well-educated students that are in their professional lives [now].
So what are you looking for when you do it? It was started by ... give a little short history of Xerox PARC. It started off as a XEROX research facility. Many big companies has research facilities.
Right, I can do that. So we’ll start in 1970. It was founded on the West Coast, which is far away from where the headquarters were. And it started with a sole mission, and the mission was to invent the office of the future. It hired only generations of computer scientists and other scientists to really think about what the world would look like and what the office would look like in the future. And a lot of the early work, the kinds of things that you mentioned earlier, came out of those first formative years. The first decade or two, the graphical user interface, the laser printer, the mouse, and ethernet and the GUI object oriented programming. Those are the kinds of things …
That many other people took advantage of.
Now this notion of kind of connected worlds seems to be ...
Yes, it started there. And it famously had its ideas stolen by Bill Gates and Steve Jobs, who then later went on to be Bill Gates and Steve Jobs.
But I think the mission is, your question about what we’re looking for today, is the same thing. What we provide to the world is we become partners with our clients to really envision the future for them and with them and really create a set of technology options, if you will.
Right. No. A lot of companies have their own research. Microsoft has an enormous research. So does Google. So does ... but most companies don’t, right? They contract with you? Is that how you’re paid? Can you explain the business plan of PARC?
Yes. You would be surprised if you actually looked at our set of clients, how big their internal IT budgets are.
They already have.
They already have. They still come to us, and for a number of reasons. So you asked earlier kind of our, where’s our funding coming from, right? So we’re still a subsidiary of Xerox, so slightly less than half of our funding comes from Xerox. And then the rest is our open innovation business, and that’s where we work with government agencies or commercial companies. So, again, they’re kind of different in the nature of research that they’re seeking for, but essentially it’s about the future. It’s about what might the future look like, and here’s a tough problem in a particular domain. It requires deep technical and scientific investigation to be able to solve it, right? And that’s what we provide. We put our brains together and they look at that problem and they provide potential solutions or options.
But it’s not just ideas about what might be the solutions to that. We do the deep science and the scientific investigation, but we do put prototypes together. We provide them proofs of feasibility. We provide them proofs of concept. And we do that in parallel with our strength on the business side, so the business modeling side, the market strategy, go-to-market. So essentially we not only deliver then the technology options, but we also deliver them how to think about really coupling that with the right business model and with the business strategy.
You’re trying to give them a full menu of ... if they’ve decided to create the best chair, whatever the heck you’re working on.
And do you tell them what you’re working on or do they say, “We have a problem with ... blank”?
It goes both ways, I would say. So there are emerging areas in technology, IOT, artificial intelligence, machine intelligence, autonomy, digital manufacturing ... This is some of the areas that we are very actively pursuing. And they’re interested sort of generically in those areas. We give them the kinds of capabilities we have, the technologies we have developed in the recent past. But sometimes they come and say, “We have this problem, and how can we go about that?”
And then you get paid for that, correct?
You get paid for that. Can you say what your budget is?
We cannot, we don’t disclose that.
Right. But how many clients do you have?
It changes. It depends on the duration of the projects that we have, but we have a very large commercial client base, and we work with all the major government agencies.
Okay, you mentioned a spate of topics that I want to get to in the next section, in talking about automation, robotics, manufacturing, just-in-time manufacturing, all kinds of things, and what you guys are working on and where you see it going.
We’re here with Dr. Tolga Kurtoglu, who is the CEO of Silicon Valley research and development company PARC. It was previously known as Xerox PARC. And we’re going to be talking about where research is going and where the future is going, I guess, when we get back.
We’re here with Dr. Tolga Kurtoglu, the CEO of Silicon Valley research and development company PARC. We’ve been talking about the history of PARC, which is very famous in the history of Silicon Valley.
But I want to delve into some of the things they’re doing. Tolga, you had been working on artificial intelligence, machine learning, and you mentioned some others. Let’s talk about each of these areas and what’s happening because it seems to me, and maybe I’m just a casual observer, that Silicon Valley’s gotten a little more serious again and they’re moving into some of these issues. You can pick any one of them. Automation, you mentioned, or manufacturing.
Yeah, I’ll give examples in each.
Let’s go through each of them. Automation.
So automation, to me, is part of, as we think about it, the mission intelligence. If I think about this, there’s three major emergent technology trends that are happening simultaneously. And actually the potential kind of comes together from the intersection of those three things. And one of them is this notion of IOT, and people use that in various different ways. But essentially what it is to me is the ability to embed sensing and sensors in physical devices, and this can be in the industrial context or it can be in the consumer context, and be able to basically generate data about that device in real time. It gives you a lot of advantages ...
If you have a tractor and you know how the tractor’s working or the plane or the ... whatever.
Exactly. It gives you this ability to sense and control and do that in real time, which is a very, very powerful notion.
The other one is this notion of machine intelligence, and automation is a big part of that, which we’re seeing more and more systems that can mimic and do the kinds of things that we thought only humans can do really well.
Mm-hmm. Give me an example.
Classification is a typical example in the AI, right? So if you think about looking at an object and telling whether that’s a car or not, that’s something that a 3-year-old kid picks up fairly quickly after somebody reinforces that, right? So it says, “This is a car.” They kind of get a sense of cars come in different shapes, models, colors and whatnot very, very quickly. And for machines to be able to do that was very, very difficult because you knew a lot of computation, a lot of data, and a lot of label data.
So we’re seeing advances in computation, advances in access to computation as well, right? This whole notion of cloud computing and the ability to have as much computation power as you needed is very powerful. And, to the first point, a lot of data is available today that was simply not available. And that enables us to train these, what we call the artificial intelligence algorithms, in a way that they can do those kinds of things.
So now they can act like 3-year-olds.
They can act like 3-year-olds, exactly. And that elevates. That elevates their intelligence level and that’s why we hear about smart TVs, and smart trains, and smart cars.
Presumably they’d be affected by these sensors, along with the ability for the machines to react and do things with.
It’s exactly that. So they can hear, they can sense, they can act, and they can respond to their environments in a way that, again, in an intelligent and a smart way.
So talk about robotics, which is linked to it, the same thing.
Yeah. I think robotics is, to me, it’s a domain where all of this is happening as well. It’s really interesting. So you ask what are we doing in some of these domains, AI and robotics. I’ll give you one example. Really interesting project and technology that we’re working on is about how to bring together these AI agents or computation agents and humans together in a way that they form sort of collaborative teams to go after tasks. And robotics is a great domain for really exploring some of the ideas there. So there’s a lot of ...
So meaning collaborative. We already do work with computers.
We just don’t respond. We just tell them what to do and they bring, pop an answer from Google or whatever.
That’s precisely right. So you can actually prescribe or ask or inquire an answer from a computer. What we’re talking about here is more of a symbiotic team between an AI agent and a human in a way that they solve the problems together. It’s not one of them tells the other one what to do, but they go back and forth. And they can formulate the problem, they can build on each other’s ideas and things of that nature.
And it’s really important because we’re — as you mentioned, Kara — we’re seeing significant advancements in penetration of AI technologies in almost all industries, right? So the more there’s going to be automation and AI in those ... you know, they predict which stocks we should buy to which songs we should listen to, when our Uber service will arrive to everything in our everyday lives. And you can kind of stretch that to the extreme. At some point, there’s going to be a huge issue with people really taking the answers that computers are suggesting to them without questioning.
So this notion of trust between the AI agents and humans is at the heart of the technology that we’re working on. We’re trying to build trustable AI systems. And one way to get there is for the AI systems to be explainable. So imagine an AI system that explains itself. So if you’re using an AI agent to do medical diagnostics and it comes up with a seemingly unintuitive answer, then the doctor might want to know why, right? Why did you come up with that answer as opposed to something else? Today these systems are pretty much black boxes, so you put the input, it just spits out what the answer is.
“You have cancer. Thank you.”
Yes, right. Exactly.
“Have a nice day.”
What we want to enable is for these systems to explain themselves in a way that they ...
How they thought of it.
Yeah, how they thought about it. Here are the assumptions that I made, just like we explain to others here are the assumptions that I made, here are the paths that I’ve considered, here are the paths that I’ve ruled out, and here’s why.
Why do we need explanations? They’re smarter than we are.
I mean, that’s the problem, right? So, again, I think that as you push the boundaries of automation, I think it’s likely ... well, two things. As you push the boundaries of automation, it’s likely to hit the barrier there in which everything is controlled by these things, but that’s ...
Well that’s the fear, right?
Societal, it’s a societal kind of interesting question.
But don’t they already do that? Don’t we already rely on them for maps now?
Just think of when’s the last time you looked at a map, figured it out for yourself. Not ever.
Yeah, so it seems a distant past. But I remember. I remember very vividly when I first moved to the U.S. I was trying to do online shopping for travel and I was supposed to get my credit card information. And I was like, “Should I really trust this?” because it wasn’t the common way of booking.
Right, but then you did.
Then I did it. I think with time, as the adoption curve goes up and whatnot, there’s a certain level of trust that builds in, but I think the rate at which these algorithms are advancing now, they’re going to be everywhere. I think it’s going to be significant.
They’ll be talking to you, explaining to you, like regular people. They’re gonna be ... but not rude, for example. Just give you nice answers and stuff like that.
That’s what we’re working for.
You could make them a little rude in different places.
We’ll talk about that, the societal impact and the responsibility of Silicon Valley in the next section, but what about self-driving? Are you guys doing stuff in self-driving cars?
Well, we are indirectly. We’re not developing, I would say, our own version of a self-driving car, but there’s subsystems that then would support the autonomy that we’re working on for the sensing technology, for example, the radar technology and whatnot. There’s a number of areas in which we develop new technologies that would then enable better sensing of the information that comes from computer vision and some of the work that we’ve done in metamaterials.
That’s gotta be the most tricky problem of all, in all computing, so I think.
It is one of the most interesting.
So many inputs all at once. Everyone’s like, “Why don’t these work?” I’m like, you know how difficult it is to drive a car? Merge, just anything.
And again, I think cars are really interesting examples. From the previous conversation as well, fully autonomous versus ... and I talked about this recently at a keynote ... It’s great. You can have a self-driving car on a perfectly sunny, California day on a highway that’s well marked and everything is kind of consistent with how you modeled the world. But take that same car and put it on an intersection in a different part of the world — you know, Asia, where it’s highly densely populated and traffic flows in a very, very different and chaotic way potentially. Or put it on a winding road, which is foggy, in the countryside.
I’d still trust it more than my mother, but okay.
That should be fine. But again, I think there are these cases where ... because ultimately the learning is statistically based. The more data you see that represents the real world, the better you can get.
And people feel more comfortable in these situations as time goes on. I mean, that’s the thing. At first you’re discomforted and then a second later you’re using it, and then the third later, you insult it. Like, “It’s not fast enough. It’s not enough,” kind of thing.
Yeah, I think the fast and the accuracy, those things are improving really rapidly and radically. The point I was trying to make is there’s still the complexity in the world. There’s still those edge cases, if you will, or cases that are very, very complex, where humans are really good intuitively or experience based to kind of look at the situation at hand and make a judgment, where computers can’t. And that’s, again, an area that we’re working on. That’s what collaboration means, in addition to what I talked about earlier.
And what about manufacturing? You mentioned manufacturing.
Yeah, manufacturing is a huge area of interest for us.
Because I think it has an enormous potential for disruption. Manufacturing is going through the digital transformation, if you will, that a lot of other industries have already gone through and completed to a great extent. But if you look at the world of manufacturing, there’s a lot of things that are still being done, either manually or that are things that are still being done that even by expert and tribal and sort of tacit knowledge, if you will. It is, I think, moving in a direction where — with the emergence of these computational techniques and AI and data and other things — that it is on the verge of digital disruption, and that’s what we ...
So explain what that would mean. Give me an example.
Sure. So, I’ll give an example from one of the technologies that we worked on recently. For example, if you want to get a part made, manufactured for you, typically you have to send out a request for code and you send to a bunch of suppliers, and they need to kind of look at that. They have experts that understand the complexity so that they can retool it. “Do I have the right machine, right resources, assets, etc., on my production environment that I can make that?” So this process is a back and forth between design and manufacturing, which has sort of an artificial boundary in between, if you will. The designers specify what needs to get made, and then the manufacturing engineers really need to figure out how to get it made, right?
So that translates into delaying to go to market, delaying the productivity of the machine shops, and a bunch of other inefficiencies, if you will. So we developed a platform that kind of brings those two worlds together. It’s a cloud-based service platform that really maps designs with supply chain capabilities.
So you basically say, “I want this coffee cup to be made,” and you can drop that on the website and it understands manufacturing processes through advanced AI and modeling and simulation algorithms. So it can actually look at the design and look at, if given a shop, it can use a virtual model of that shop with the processes and simulate how you would translate that design into machine operations or instructions.
Right. Without the back and forth. Just instant.
Without the back and forth, right, in real time. And you can say if you were to go with this particular shop, this is how they would do it. This is how much it would cost, and how long it would take. If you don’t want to go with them, here’s another alterative, and then a different set of machines, different set of people, a different set of processes, what have you. It’s a different answer. So it really digitally connects the supply, the industrial base, which ...
Which is now done by flying to China or ... you know what I mean? People, they just do that, which seems insane.
They do, and they’re experts that really understand what you’re trying to do and understand the supply chain, and really connect that, right? So in this particular example, we build a platform that kind of brings that together for practical domain.
Okay. When we get back, we’re here with Dr. Tolga Kurtoglu. He’s the CEO of Silicon Valley research and development company PARC. We’re talking about innovations and where they’re going. When we get back, I want to talk about the responsibility of Silicon Valley to these innovations, and also some of the way-out ideas he thinks that are happening, because that’s what PARC is known for.
We’re here in the red chair with Dr. Tolga Kurtoglu, the CEO of the Silicon Valley research and development company PARC. We’ve been talking about a wide range of where things are going, including artificial intelligence, machine learning, security and other areas.
So let’s talk about ... you’re talking about constant innovation, the business of innovation, which I think is the business of Silicon Valley, trying to be. How do you assess where Silicon Valley is now, because these areas of innovation change and morph over time. Obviously a lot of people are worried they’re shifting to China or somewhere else. Probably China would be the biggest choice. How do you assess Silicon Valley’s health right now, in terms of innovation?
In terms of innovation and innovation leadership, I think Silicon Valley continues to be the world leader in innovation. But you’re right. I think the pace of innovation has changed dramatically. And to a certain extent, I make an analogy to kind of riding waves. So these innovation topics and areas come in waves, and when you’re there in the water, you kind of have to catch on to that wave, stand up and ride it, and then you have to think about what comes next and whatnot.
So I think in terms of really defining what those waves are, Silicon Valley holds the leadership today, as we see in the Internet of Things, AI, self-driving cars and a bunch of other areas, robotics for sure. Still the majority of innovation comes from the Valley. The center of gravity is closer to our coast than other parts of the world, but clearly there is competition elsewhere in the world, right? So there’s pockets of areas and regions that either aspire to or that already have a solid base for innovation and technology development. I think partly the continued leadership is continuation of a number of factors that kind of made Silicon Valley Silicon Valley in the first place. And those factors are still there.
In my mind, there are three or four, right? So people often ask, “Such and such place would be the next Silicon Valley,” or “We want to be the Silicon Valley of X.”
There will be an X Silicon Valley.
Sure. But I think it comes down to four factors in my mind. There is access to capital, which we have with the VC community and otherwise. Access to talent, with not only one of the two best universities, but more in all the world’s talent want to be here in the Valley. But also access to what I call entrepreneurial mentorship. That’s something that people often overlook. We had fair trials, and if you go back to the history of Silicon Valley, a lot of those people that were influential in the formation of the Valley and the silicon industry, they moved on, and they themselves became investors and mentors. And they educated the next entrepreneurial generation. And that kind of continues.
So there is a tremendous amount of know how to work in terms of entrepreneurialship, and how to really think about technology and innovation and take it the market and make it a success. That’s hard to repeat elsewhere.
And then there’s a really interesting connection, which is the fourth factor, is the people of Silicon Valley and the culture here. Often when we talk about innovation, we talk about early adopters and laggards and the ones that follow, right? So Silicon Valley is full of people that are early adopters, that are dying to try out the next big thing, right?
So if you think about all the technologies that we use today in our daily lives from Waze to Lyft and Uber and other things, you know as they come out, there’s already a cultural test bed, if you will, that are willing to really try out those things.
All right. So that’s the happy case, but some things are starting to show. What’s happening at Uber is ugly.
And sort of a nasty version of what’s happening. It’s arrogance. It’s sort of the end of an empire kind of behavior. It’s messy and sloppy and not at all these vaunted ideas.
What’s the danger to innovation in these cultures? Obviously, money. It’s warping.
It generates a tremendous amount of wealth for a limited number of people. And then it comes down to individual preferences and character as to how people deal with that, right? So that’s certainly a risk. And I think there is also a danger of, with great power and leadership comes also responsibility. So you really need to understand what that means. You’re changing the world, you’re really disrupting major industries in profound ways. And you can behave in a way that then shows a certain character versus you had behaved in a way that shows you do that with a certain humility.
So, again, what I think we’re seeing is, in the examples that you’re saying, that there’s a potential to kind of be too arrogant about all of that and behave in ways that is not very productive for our culture.
Right. So talking about great responsibility, do you think people in Silicon Valley do understand the impact of their ... I had a really interesting meeting yesterday with someone from Facebook who — we had a large argument about their responsibility in the election, and obviously they were grappling with it. They’re definitely concerned about what their impact is.
Same thing with the self-driving car stuff. What happens to jobs? Don’t you understand the linkage between job elimination, or Amazon and retail? You could iterate through lots and lots of innovations that are happening. And machine learning and automation, they all have a profound impact on our society. And you can see it politically right now, the rise of populism, the rise of really idiotic leadership that takes advantage of those fears about the future. How do you look at that? Or do you think Silicon Valley has to do something to understand its responsibility?
I think it does understand, because there’s not only the sort of the bad examples, if you would, but there’s a fair amount of good examples that are coming out of Silicon Valley as well. I mean, through either philanthropic causes or some of the discussions that are going on. You know, off the top of my head, I’m trying to think of examples.
When you think about robotics and AI and automation and jobs, this whole notion and conversation about universal basic income, taxing robots and AI agents, and really kind of creating value for the society at large, I think those discussions are happening. We know of research centers that come together that have a social mission in trying to use artificial intelligence and machine learning for those causes and for those causes only.
So, again, this is not black and white in my mind in terms of Silicon Valley is all bad or all good. There is a subset of people that really understands it comes with a certain level of responsibility, understanding the impact of what technology development means, what innovation means for the society here, but also for the world at large. And there’s some people that simply don’t think about that, or choose to act in a different way.
I think this is the first time they really understood jobs, especially job elimination, because some of these technologies ... It’s not a photo app. It’s not ... they sort of decimated media certainly, but nobody, they didn’t seem to care that much about that.
Now it’s getting to the heart of a lot of things. When you’re talking about automation and self-driving cars, you really move quickly through a large swath of the population and how you ship them.
We just had an interview with Reid Hoffman and Marc Andreessen, and I just did another one with Reid. I had a really interesting back and forth with Marc Andreessen, who’s one of the most prominent venture capitalists. And he was saying, “It was better when we got to manufacturing over the farm economy because there was more jobs than ever.” And I said, “Yes, but the transition.” And he was like, “Well, there were blacksmiths and then there weren’t.” And I was like, “Yeah, but what about the blacksmiths’ family? What about the people they bought food from? What about ...”
I think there was a great deal of social discomfort during that time that we utterly forget about, but was very profound at the time. So who is responsible for that from a technological point of view, because here you are saying, “We’re going to make this of the future, that of the future.” What about the present? How do you deal with the present? Would you think about that? Is it just resurgency, or “Just forget it. We’re just going to invent time machines and invisibility cloaks and whatever happens, happens?”
We’re all individuals and we think about what that might mean for the future because absolutely we want to impact the world in a positive and significant way. And again, with that comes a certain level of responsibility. You have to kind of think about, “Well, those are the positives, but what about the negatives?”
And yes, I think this is an interesting moment from that perspective, with all the advances in automation and AI and robotics and the kinds of things we talked about. There is the reality of jobs being lost and certain people being displaced from the workplace in a profound way. And I think that there’s two pieces of perspective on that. One, we need to think about also technology and what opportunity it creates. So it takes away certain jobs, but it also presents opportunities for different kinds of jobs, different kinds of skill sets to be significant.
Let’s talk about that finishing up. What do you imagine ... I’ve seen recently, I’ve been reading a lot of stuff around what the work that’s needed in the future is, what the actual jobs are, and what’s going ... A lot of them are eldercare, childcare, things that are pretty basic. What do you envision will Xerox PARC be looking at in 20 years? What are the things ... Right now we’re doing that automation and AI kind of thing. If you’re thinking way out, what is it? Like time travel or telepathy or what?
No, I don’t know if this is going to be way, way out there, but I think the answer is kind of an intersection of those technological trends that are emerging. We think about the office place of the future and how that may look like, right? So it’s most likely going to look very different than the one that we have today.
And it’s also very likely ... If you look at the consumer space and how people interact with technology. There’s a slew of sensors on my smartphones and other things, variables in our smart watches, etc., and all data that is coming from that is sort of contextually processed. And I think the future is a bit where we all have as individuals these systems, if you will, that would then help us with what we need to do. Like if you’re trying to find food, we can immediately get recommendations about that. If you want to make reservations at the concert and whatnot, they can get done. It’s like these virtual aids and assistants surrounding us in real time and processing a lot of data that are coming from different sources to kind of really help us out.
And I think the future of work is no exception to that. It’s going to be really interesting experiencing that ... The office spaces, whether similar to what we have now or radically different in different places, would be loaded with sensors. It would be loaded with machines that can talk to you and that can interact with you. You can think about how we have the Amazon Echoes and Google Homes at home and interact with them. And all of that data and all of that interaction would be delivered to people in a way that those agents understand the workflow, the task, the corporation hierarchy and those kinds of things, and able to deliver you personalized experiences, given the task that you want to do.
Yeah. You are talking about a fully monitored society.
That is the risk. And then another area we’re working on ...
No, that’s exactly what it is.
That’s exactly what it is, right? And then I think one of the biggest question marks is what does that mean with respect to security and the privacy of people, right? And that’s another area ...
There is no such thing, right? I’m sorry. I thought that was gone 20 years ago.
It is a significant challenge, yeah.
So what does that mean? Obviously we’ve seen cyber attacks. Let’s finish up on that. We’ve seen all kinds of threats, the more we depend on these technologies.
Yeah, I think my prediction is it’s only going to go up, right? So, we need to think about the signs in technology that would then make our data, our workflow, more secure and more private as well as the personal data more secure and more private.
Financial data, other personal identifiable data and sensitive data and whatnot. But also, on the physical or the cyber-physical systems as well, right? So if you have an autonomous car, you don’t want somebody to take control of that autonomous car and do whatever they want with it, right? Same thing with the drones and whatnot. So it’s really fascinating if you just kind of go beyond the traditional notions of security and privacy with the data, but we have all these smart systems and machine intelligence, I talked about, and all those systems need to be secured as well.
Yeah. Because humans never use good things for bad purposes.
I had an ongoing fight with the Facebook people about their Facebook Live. And when they first showed it to me, I said, “Oh, someone’s gonna kill someone.” And they’re like, “Kara, you’re such a bad person.” I was like, “What? Am I the only person that’s thought of this?” And obviously they had thought about it, but it was my absolute first thought, not my ... And they were like, “It will be wonderful.” I’m like, “It will be awful.” So it’s a really interesting ... I have a terrible perspective.
But can we finish up? Let’s finish up. What is the craziest thing that you’ve seen lately, like there’s something that you’re like, “Whoa, that’s pretty cool.” Because you couldn’t have imagined the graphical user interface, or even at the time that was probably like, “Whoa, that’s pretty cool,” because the way computers and we interact, and no one ever thought about interacting with them in a friendly way. So what’s the thing you’ve seen and you’ve gone, “Oh my God?”
It’s a really interesting question.
I know you’ve seen something.
Yeah, well, I have, but this goes back to sort of being immersed in technology so much that sort of the surprise element or the wow factor, if you will, is just kind of gone.
Okay, I’ll give you an example. When we had Regina [Dugan] when she was heading DARPA, she started talking about like a Mach 10 plane, or something Mach whatever that went across the world in 30 seconds. That was “Wow.” That was “Whoa.”
Yeah, I’ll give one example.
It’s a problem for humans to be in it, but that’s another issue.
I’ll give kind of ... We talked about security and privacy. I’ll give one example from PARC. When I first saw it, I wasn’t part of that group. I was like, “Wow, that’s really cool.” And that’s a self-disintegrating electronic circuit, which is basically built in a particular way on a particular glass base that you can then basically put a lot of energy in, and then you can remotely release that energy, trigger that. And then when it releases, it completely disintegrates.
What? Why does it disintegrate?
Well, you can imagine that there’s electronics in which there is data that you don’t want anyone else to have access to.
Oh, so like “Mission Impossible.”
It was exactly my reaction. I was like, “Wow, that’s really cool.” That’s a cool technology.
So it would just disintegrate on command, like whatever.
On a trigger. You can do it many different ways.
You don’t have to like break the cellphone dramatically. It just would like ... Or have it be a Samsung Galaxy or whatever.
Exactly. Imagine your phone and you know where it is and you can remotely just trigger that, and then it just kind of goes away.
Wow. That’s interesting. No time machine then.
No time machine here.
I’m waiting for someone to come here and tell me about how the time machine is going or the invisibility cloak or ...
I would be happy to come back for that.
Well, I’m waiting. All right, Tolga, it was great talking to you. Thanks for coming by.
This article originally appeared on Recode.net.