ACM ByteCast - Alvin Wang Graylin - Episode 63

Episode Date: January 30, 2025

In this episode of ACM ByteCast, Rashmi Mohan hosts Alvin Wang Graylin, Global VP of Corporate Development at HTC. Alvin is also Chairman of the Virtual World Society, Vice-Chair of the Industry of VR Alliance, and President of the Virtual Reality Venture Capital Alliance. Alvin is a leader in the virtual and augmented reality industries. As a serial entrepreneur, he founded four venture-backed startups across sectors including AI-driven conversational search, mobile social networks, ad tech, and big data AI analytics, spanning both China and the U.S. As an active investor, he funded more than 100 startups and played a pivotal role in the establishment of HTC’s ViveX Global VR accelerator and SOSV’s mobile internet investment incubator. A frequent keynote speaker at leading international conferences, Alvin is regularly featured in major media outlets, where he provides thought leadership on immersive computing (AR/VR/MR), AI, entrepreneurship, venture investing, and the Chinese market. His latest book, Our Next Reality, examines how AI and XR convergence will reshape society. He holds an M.S. in CS specializing in AI from MIT, an M.S. in Business from MIT’s Sloan School of Management, and a B.S. in Electrical Engineering from the University of Washington, focusing on VR, AI, and CPU architecture. Alvin describes his early introduction to technology and computers after immigrating to the U.S. from China and later working at the first VR-focused research lab outside the military at the University of Washington. He highlights some of his pioneering contributions to the consumer PC, mobile, and internet industries in China, including helping establish Intel’s China office. Alvin explains how his background in NLP and AI helped him navigate and address the complexities of Chinese as a language for mobile search and shares some milestones from his work in China, at companies he founded and later at HTC. He touches on the early challenges and limitations of VR, progress made, and its exciting future promise. He also discusses some major advancements in XR technology, its transformative potential for education, and explains the synergy between AI and XR and the need for positive adoption at this pivotal moment in technological history.

This is ACM ByteCast, a podcast series from the Association for Computing Machinery, the world's largest education and scientific computing society. We talk to researchers, practitioners, and innovators who are at the intersection of computing research and practice. They share their experiences, the lessons they've learned, and their visions for the future of computing. I am your host, Rashmi Mohan. The best way to predict the future is to create it. Well, our next guest is a groundbreaking technologist, serial entrepreneur, investor, and celebrated author, and has been creating the future for many decades.

Alvin Wong Graylin is the Global VP of Corporate Development for HTC and has previously served as the China President at HTC. He has founded, built, and successfully exited from four startups in the mobile world and is usually found at the cutting edge of all things technology. He is the President of the VR Venture Capital Alliance, Vice Chair of the Industry of VR Alliance, and now the Chairman of the Virtual World Society, following his words with real action. He was also recently inducted into the XR Hall of Fame. He is a Distinguished Professor at Bayhang University and a recent author of the very well-received book, Our Next Reality. We are so thrilled to have an esteemed technologist on the show today.

Alvin, welcome to ACM ByteCast. Well, thanks for inviting me, Ashley. I'm looking forward to our conversation. Likewise, likewise. So Alvin, I'd love to start with a simple question that I ask all my guests. If you could please introduce yourself and talk about what you currently do, and maybe give us some insight into what drew you into this field of technology and computer science. Sure. So as you kind of just said, I'm doing quite a few different things, both in terms of industry as well as corporate as well as more for cross-industry organizations.

And my focus really right now is to see how we can make technology something that does good for for humanity and for society versus purely for profitability as you probably mentioned earlier i've been involved in studying and developing product and services and building something for the better part of three decades and i guess what really got me started was when I first moved over from China to the United States and my father sat my brother and I down and said, hey, you know, we brought you over from China so that you can come to America with the best education in the world better than when you came or before you came. And you need to find what it is that you can do to maximize that goodness. And I kind of gravitated towards software and technology and tech in general, just because when I first got here, one of the first things I did was with the money that my brother and I earned delivering papers and working five jobs is that we bought ourselves a computer. And from that computer, we started to do programming and get online and go on BBSs. This was back in 1980, 81.

So it was like Commodore 64. And then later on, we had an Amiga and then later on the Atari 400. So just really got gravitated to the computing and tech space and online medium, because really allows an individual with no other tools except this computer to start creating things and to be able to see fairly immediately the impact of your work. And for a young person to have that level of access, I think I felt very fortunate to have been given that opportunity by moving to a country like the US. It's amazing, Alvin, that you were able to enter the field. I mean, I think it's particularly impressive that you and your brother were earning money and then sort of providing

for your own interests. It's also amazing that your father had the vision to kind of sort of give you that freedom of expression and to tell you about, you know, I think it's such a powerful thing for a lot of us to know, you know, young or old, that we all have a gift that we can give and leaving the world a better place than where we found it is so critical. Did you always think that, as you were sort of, I'm guessing, very self-taught and learned a lot of computing on your own, did you always think that that's what you would study in college? Was that just sort of a natural progression? Well, I was doing essentially tech pretty much since about nine or 10 until college. In fact, in college, I was building and selling computers and doing custom software and network for small businesses

and the university to pay for a part of the college. So it was something that I was pretty good at and something I enjoyed. But to be honest, I was actually deciding between going into EE or going into aeronautics because I was in Seattle and Boeing's based here. So aeronautical engineering was also an area that I was considering. And the funny thing is when I was deciding between going to IBM for an internship to work on what became the PowerPC chip or going to Boeing and working on the B-1 bomber and talking to some

of the folks at Boeing. And they said, oh, you know, if you come here, you'll be in a room where there's no windows and you'll never see the sunlight during the whole time you're here. And you can't tell anybody about anything that you've ever done, even after you leave. So after hearing that, I was like, I think I would rather work on something where I can actually see the sun and also be able to have conversations outside with my colleagues and friends and and be able to to enjoy understanding things and discussing things so so I kind of moved over towards the the double e side and which also brought me later on to intel and then but at the

same time you know I was I was also working at the Human Interface Technology Lab at the University of Washington, which was the first VR-focused research lab outside of the military. And I did not come for this, who was actually just the former chairman of the Virtual World Society. So I just took over for him a couple months ago. So I was kind of going full circle to see the cycle of life, I guess. That's incredible, right? I mean, that's a long association. It sounds like Tom Furness was one of your mentors when you joined the Human Interface Technology Lab. I think that was at University of Washington, if I'm not mistaken. Yes, yes, exactly. Yeah. I mean yes, exactly. I think it's important for any young person to have mentors that

care about you and really can give you useful and unbiased advice. Right. So we should all try to seek those out because just a few people in your life can really change the trajectory. Absolutely. Yeah, absolutely. Including the one that told you about the windows also moved you in a different direction, even if it wasn't towards them. I'm also curious, Alvin, the field when you entered it, just in general VR, was super nascent, right? When you started any serious work on it, that's amazing because you have sort of a green field for discovery. But I'm sure there's

probably a lot of challenges also that you may have encountered, because it's an area that nobody has sort of really explored seriously before. Can you talk about both? Like, what did you find exciting? What did you learn through that experience, as well as what were some challenges? Yeah, I mean, I think, you know, going into any industry, and I've had the luck of being at the beginning of multiple technology kind of revolutions over the last four decades. So, but you know, the VR space, when I got in there, it was really the early days and the devices we were working on were hundreds of thousands of dollars. So they were definitely not consumer devices and you had a lot of wires and

there was two and a half kilograms on your head. So more than five pounds of these devices, whereas today you can get them for a couple hundred grams or some new, you know, the eyeglasses or XR glasses can be under 100 grams. So the weight, the comfort, the resolution, everything has just dramatically gotten better. And the best part is also probably several hundred times cheaper than it used to be as well. So it's definitely a lot more accessible.

And one of the other, I guess, challenges in addition to comfort is also just the software. Because the processing wasn't there, you're more likely to be nauseous. So you have to kind of get your VR feet, as they call it. So the more you use it, the more you get used to it. But I think that's kind of on par for any new technology. Whenever you're getting into the early days, you have to deal with the deficiencies or immaturities of a new technology. I was also very early in the internet space

where when I joined Intel was when the first browser came. And to see that entire industry come up. And when I went to China, I actually helped to create the consumer PC business there and seeing, you know, people start having PCs at homes where, you know, in China, there was only business PCs that was two to $3,000 when the average income of a projava was something like two or $300 a year. So this is kind of 10x per yearly salary just to buy a computer. So I was able to help create the various things that fill in the various things that was missing in industry in terms of lower cost devices, lower cost channels, as well as localized content to help feel the growth in that business and the growth in that use case. And by the time I left, we went from

zero to about 300 in chips being sold into the consumer PC space. So just three, four years later. And then later on, I was able to be part of the entire mobile revolution as well, where when I went back the second time to China, I started the first mobile search engine in China. And at that time, there was no mobile internet. So we were doing text messages to do conversational search. So kind of like what GPT searches today, we were doing back in 2005. So almost 20 years ago, we were allowing people to send a message, send a question and have a response directly versus a link because this was before they had WAP or apps or mobile or smartphones, really. So, you know, to see that grow and then to have seen the adoption of 3G and seeing apps coming into mobile phones and then getting to that side of the business. And then over the last nine years, seeing the growth of the XR business from

essentially something that was disappeared for about 15 years to coming back onto the scene and being able to restart something that I had been thinking about for multiple decades, to get back into that industry and then play a role. I feel just very fortunate to have been in all these right places. There are so many things in there that I want to talk about, Alvin. First thing being, you've had multiple stints and from what I hear going back and forth to China, was how do you spot those opportunities? Was China a natural choice because that was home, because that's where you came from? So that's maybe my first question. And I have like follow up questions as we progress. Yeah, I mean, I guess I was lucky in

the sense of when I went to Intel, the first time I went back to China was to help found the China office for Intel back in 1994, exactly 30 years ago. And the fact that I spoke Chinese and was from headquarters and had an interest, they made me an offer to go there. So in a way, it happened to me more than I chose it. But the fact that I was in that position, it also gave me an ability to execute a lot higher level of things than I would have normally at my age. I was 23 and managing a business of multiple hundreds of millions of dollars and hiring dozens of people or more, or actually later on, computer people who I was managing

and driving and sending direction for, all of whom were older than me, were better educated. And the only reason was I was lucky enough to be a Chinese speaker who was coming from headquarters who had that interest and was there at the right time. So sometimes things happen because you're lucky. But then if you can take those opportunities and really show your ability to execute and your ability to analyze the situation and bring value, people will continue to trust you to do more and bigger things. Absolutely. I think that's a very valid point that you bring up, which is ability to execute. Opportunities, of course.

One is to be able to spot that opportunity, to be able to make the most of it, right? The responsibility that you're given, of course, it's a super young age to be able to sort of have that level of responsibility over a business. But it's amazing that you were able to, you know, capitalize on that and learn and contribute positively. While you were in China, I know you mentioned that you were also a serial entrepreneur. I know you were in mobile search and advertising. Again, I mean, it seems like a pattern again, brand new area, you know, mobile technology was

probably just sort of burgeoning in the world and in China at that time. What was your experience of building in sort of a brand new sort of technology area? Were there like, you know, learnings from your both your previous experiences, as well as your education that you felt were like a continual thread that you could apply and build on these on this new sort of new world of mobile technology? Yeah, absolutely. And, you know, we're all really a product of our prior experiences. So part of when I was actually at the University of Washington, I was studying natural language processing and working on neural networks and AI back then. And later on at MIT, I also did

a lot of symbolic related AI. So that type of experience played a big role in terms of helping me come up with the right solution set to enable us to actually provide workable solutions, given the limited processing that we had and the difficult complexity. In fact, with Chinese, it's actually a more complex language to work with because there are no spaces between words. So there's a lot of different ways that you could cut a were very different things. So creating the right context to create proper natural language is probably more difficult than it is in the States. And there's a lot more to work with. But, you know, so that experience was definitely, from a technical perspective, helpful.

The other thing was really just the experience of having done a startup after MIT in Boston and working through the difficulties of trying to sell to large companies, trying to get the big portals and e-commerce companies to work with us. When I went to China, I had to do the same thing, but to state-owned enterprises, to large carriers. So I was able to, after we built our solution, sell to Chinese carriers and have them implement

our search solution in their data centers and have our staff operating within their premises to get these services up and running. And so that kind of prior experience definitely helped to feel and make me prepared to take that on. But again, when you're working in a brand new space, there's always new difficulties that you don't think of. Really, I think finding good partners and good teammates that will fill in for things that are missing on your side, because there's no way a single founder is able to have the total solution.

Absolutely. I mean, I know I was listening to some of your older interviews, and I know that you had a really, really sort of successful set of partnerships. I know that you had a really, really sort of successful set of partnerships. I know you were a partner at the Olympics. You partnered with China Telecom. The Olympics is something

I'm actually quite proud of because at the time when the 2008 Olympics came around, everybody wanted to be a partner and they were paying tens of millions of dollars to name a logo and be able to use it.

But for us to actually have done it completely organically, where we said, we think we can help you make the service and make it multilingual, make it natural language, make it work on any phone. And they kind of didn't believe us. But I had my team work for six months dedicated to that project. And we won the bid and was able to be the official mobile search provider for the Beijing Olympics, which was a real proud moment for us, even though we didn't really get any money for it. But the sense of gratification of being able to serve your country and to serve such an important entity as the Olympics was something that was worth it for us to put that effort into. Absolutely. And just the experience of working at that scale at such a global world stage, it is phenomenal. And obviously, it stuck out. When I was listening, I was so impressed. And I was like, wow, that's such an amazing achievement for a young startup company. One question I had,

Alvin, about the companies that you built, it seemed like you were building for the hyperlocal market, right? You were building for the Chinese market. Did you take these products into different international markets? And what was that experience like, like building in China, bringing it maybe west or wherever else that you took it? Yeah, so for the services that I did in China, actually, they were pretty much for the Chinese market because it was a very local language, local market. And the good thing is China is actually a large enough market where you can do that. Probably the only other place you can do that is in the US. These are essentially the two largest indigenous markets that you could operate in. And to be honest, I think Chinese companies, particularly software companies in general, have had a really tough time getting products to be accepted outside of China. I was in a period where really there was no Chinese apps or services or even websites TikTok that came out of China. But really, that may be one of the very

few that was able to cross borders and be able to have that impact. Other things like more appliances and digital equipment and so forth, those have been very popular out of China for a very long time. But in terms of the software side, I think that the cultural and language barriers and the customs and expectations and user experiences of different cultures has made it a little bit more difficult for China. But I don't think that that's going to be an issue longer term.

Although now the bigger problem is probably more geopolitical barriers that are being put up to limit the success. I think recently they've just announced that they're going to have a ban of TikTok that will start in January 1st. So these are the kind of things that we would like not to see.

I'm actually not a big fan of social networks that are not bring more innate value. And I think TikTok is probably one of the ones where you can spend a lot of time, but not really learn much. So I think TikTok is probably one of the ones where you can spend a lot of time, but not really learn much. So I'm not a big fan of it, but I think the fact that we are operating the way we are right now in the US, where you have this, essentially a ban, a particularly good reason, considering that they've really followed all of the policies and expectations of the local law, right. So for America to be positioning itself as a democratic place of freedom and free speech,

that's a little bit hypocritical. Fair enough, fair enough. And I think your point about TikTok is well taken by me. I know my teenagers will probably object or protest. But yeah, no, I hear you. ACM ByteCast is available on Apple Podcasts, Google Podcasts, Podbean, Spotify, Stitcher, and TuneIn. If you're enjoying this episode,

please subscribe and leave us a review on your favorite platform. What was also interesting, Alvin, is around this time when you were in China and you were building and working very deeply in this space of building these companies, you went back to HTC or you kind of forayed into the VR space again. How did you see that emerging interest? I'm guessing they approached you because you had expertise in this area, but what do you think kind of saw that resurgence? So it was kind of a coincidence in the sense of I had recently sold my company and then I was helping a friend start up a venture capital firm in Taiwan. And at the same time, I then got contacted by a friend who was the special assistant for Jack Ma. So he's the chairman for Alibaba.

Apparently Jack Ma was talking to the chairman of HTC and they had mentioned that they were looking for somebody to head up their VR business and if there were any candidates. And my friend who was in the room, I guess said, oh, I think I know somebody

that might be a good fit. And somehow that connection got made and I was able to go and now it was sure. So sometimes, you know, good things happen just by luck. And, you know, I think I've been the recipient of multiple lucky breaks over my career. I mean, I think it's also like an area that you were kind of one of the pioneers in. So certainly I think it was, you know, luck on both sides. I'm sure they were super lucky to have you as they were sort of bringing up their VR division. What did you think in terms of the market itself at that time, Alvin? Did you see it grow at the pace that you expected? Was it something that was sort of more

like a visionary thought from HTC saying, hey, I think this market's going to grow, let's start to invest in it? Because it was still pretty early when when you went in there i don't think it had taken over the interest hadn't taken over the world as it has today yeah so it was fairly early and it was a relatively early product because when i tested it was still the kind of the early prototypes but it's already significantly better than what's been on the market for the last 10 years in the sense of it was HTC was a company that had a room scale VR where you can walk around the room and essentially experience the holodeck type experience. Whereas at the same time, when Oculus, who was the other player in the market that was later bought by Meta or Facebook, they had ones where you were essentially sitting down

in front of a camera and you can move around, but it was essentially a sit-down type experience. So I was actually quite impressed from what I saw with HTC and that did heighten my interest. But at the same time, when I was speaking to Cher after the demo, she's like, oh, what do you think? Does this excite you? And I was like, oh, this is really exciting, but it's actually not that different than what I saw 25 years ago or 20 years ago when I first experienced this thing, because when we had those several hundred thousand dollar devices, it already had the ability to do six stops. So six degrees of freedom being able to move around. It had contracting,

it had full body tracking. So, so we, I've been able to experience it definitely at a much lower resolution and lower processing, lower fidelity, but it really was more of a linear type of a progress from where I was. But it definitely to me was possible. And even with the early ones I saw in the early 90s, I could see where it was

coming. And I had told myself, hey, within 10 years, everybody's going to be using this. This is going to be amazing. You know, three decades later, I think we're getting close to some broader adoption, but we're probably still a few years away from having this becoming fully mainstream. And when I saw it again in 2015, I was thinking, okay, now I think this second resurgence is probably we're ready to take this more mainstream. I would have expected it by now to have gotten even further than it is, even though it has, you know, gotten into tens of millions of users now. But you know, we're still far, far away from the billions of users that we see on cell phones, right? So we're still two

orders of magnitude away from where we need to be. So that's very interesting, Alvin. What do you think is sort of holding us back? I mean, when you took on the role at HTC, what was your goal in terms of like, what metrics were you tracking to see the success of this program or the product? Yeah. I mean, I was looking essentially at trying to mask the growth that we saw with the iPhone. And you look at the iPhone, the first few years grew very slowly. I mean, not very, but let's say a few million units. But by the third, fourth, or fifth year, it started to really ramp up into the tens and

then the hundreds of millions by the fourth year. And in some ways, I was kind of internally forecasting that, okay, we should be able to get into that order of magnitude and at least into, you know, tens of millions of units and then probably within five years in the over 100 million units.

So I think it definitely has not lived up to that expectation I had for myself. Although I think internally there was a mix of expectations because I don't think the devices and hardware were as ready for mainstream as I had thought it was

because it still had wires that connected. It still required high-end PCs to dock it the first few years. You know, only after the third year did we start working and relating on these standalone devices. And now we're getting to a point where I think we are really now a few years away, like three to five years away from having the hundreds of millions of devices out there. Because I think we need to move from having headsets, which are giant boxes on your head, to glasses form factors. And within the next couple of years,

you'll find that there will be probably over a dozen vendors from around the world that would be providing glasses type form factors that has displays, that has AI built in, has fairly long use time compared to less of a blue glass that I had first used back probably more than 10 years ago, those devices lasted 40 minutes and you were out of battery. And it was one eye, it was really thick and a super nerd.

Now you have glasses that look like glasses that people can't even tell unless they really look closely that these things have cameras or mics or they're what you would expect. And I think the social acceptability is definitely one of the things that held it back in the past. Now I think that currently what's holding it back is actually the lack of content that makes you want to use it every day.

What we find is that for a lot of XR devices, the first month or two, it's being used very heavily. But after that, it dramatically slows down because a lot of people run out of the content that really excites them. And so the good news is now with what's happening with generative AI and all of these AI tools that are coming out, where we're

first going from prompt to pictures and then prompt to video, and now you have prompt to 3D worlds or prompt to 3D objects. And then you have ai avatars that are fully synchronized with with voices that you can then generate to sound like anything you can essentially put npcs that are almost feeling they are alive into these virtual worlds and have it all be generated all the time and almost for free. So when you have that, we get around the limitation that we have in terms of content that resonates with you and asks you to come back.

And we will have reasons to put it on every day. And the more you put on these devices, the more use cases you have, the more data you are generating that makes you want to stay longer because now it is more useful to you than ever before.

So I think that positive reinforcing loop is about to happen. And when it does, then we will start to see the ramps that I was hoping for and expecting eight, nine years ago. Yeah, I think that's fascinating, Alvin. I mean, I think what you're talking about is also, you know, how do you sort of build that engagement? I know when the internet first came out, you know, people were thinking a lot about engagement metrics, and how do you keep people on your web page and, you know, sort of ply them with the right set of ads and the content, the time spent was a big metric that we would track. like we are at that inflection point now of greater adoption because we have now learned to understand the domains in which this is particularly valuable because I know you also spent some amount of time thinking about applying VR in the field of education. Was that sort of

just one of the areas that you were exploring? I thought it was fascinating when I heard you talk about that and I'm just wondering if you thought that the technology naturally kind of lends itself well in that domain and now we're kind of expanding way beyond that so that it's applicable to so many different areas of our life yeah i think education is one of the i think most positive use cases of this technology right in addition to things let's say like medical use case or collaboration you know gaming is kind of what a lot of people know XR for today because we're met as pushing with their kind of on your face gaming console type of a positioning for their product. Although I don't think that's necessarily the most productive or useful use case. And the reason I really lean towards education is because we as a species have evolved in a 3D space and our brains adapt best and learn best when we actually are in a 3D environment,

when we can not only see something, but we can feel it, we can walk around in it, we can let our whole body be involved in it so we can have fullbody learning. And that allows you to learn faster, to remember more, and to be able to recall it better. And I think that's the possibility where we can essentially let the next generation of learners have the ability to not just watch a movie about

history, but to actually travel back into time. Not just be able to see a document about atoms, but to actually put yourself into the size of an atom and see how they interact with each other. And that kind of visceral learning is something that you can't help but learn. You can't help but get excited about. So education definitely is an area I look forward to more use cases in.

It's been a tough area for businesses to be successful in because it's a highly regulated business or an industry. And I think there's a lot, the schools in general have been relatively slow moving in terms of their decision-making and their adoption. We even see that today with AI language models where a lot of schools and a lot of teachers are afraid of it and they're afraid of students using it because they think of it as cheating. to essentially a personalized tutor that will be available 24-7 and be able to teach them at the pace that they need to be learning at and to be able to personalize the content to them. And it should not be seen as a threat to schools if school's objectives was really to teach. It's only a threat to schools if school's objective is to make money from teaching. So I think that the education industry really needs to be disrupted a little bit and probably will be disrupted quite a bit over the next few years,

and I think for the better. And hopefully the progressive teachers who are part of that structure will enable it to happen where you're bringing in the AI to support learning instead of keeping it out as a way to preserve jobs. I think it's amazing that you bring such a positive sort of spin to the use of AI as well, Alvin, because I think a lot of us also, you know, you tend to read a lot of media, etc. There's a lot of fear, like you say, about adoption. And what does this mean for the way humans have been operating for centuries? And, you know, I kind of naturally leads me towards your book, Our Next Reality. And you talk a lot about the metaverse and just your overall, I think your dialogue in the book is

sort of really highlighting the positive outcomes of us being engaged for us embracing this technology with open arms. So I'm just wondering if you could maybe give us just a little bit of your perspective on why do you think we're at this pivotal point of adoption right now? What has changed? I know you spoke a little bit about pivotal point of adoption right now? What has changed? I know you spoke a little bit about the availability of content, but what is your sort of perspective on why we as progressive humans should really be sort of adopting or embracing this technology now? Oh, embracing AI technology or immersive technology?

I would say immersive. Yeah, immersive technology. And AI as well, I would say immersive, yeah, immersive technology and AI as well, I would say. And I'd love to hear your perspective on both because I think that there's a little bit of caution and worry on both fronts. And it's natural for people to fear change and to fear things that they don't understand fully. And the reality is that both AI and XR are technologies that is complicated to understand and to explain. And particularly, I think XR, if you haven't tried it, it's very difficult to use words to describe it to you. Because it's such a visceral medium.

And there hasn't been enough devices out there to let enough people buy it firsthand. Now, I guess the importance of why XR is going to play a greater role in our future is that I really think that the AI and XR are actually going to be two sides of the same coin. These are two technologies that are both maturing at the same time, and they're both highly reliant on each other. So almost anything in XR and the metaverse would not be possible without AI, right? Everything from eye tracking to voice recognition, hand tracking to rendering 3D objects and scanning a physical world to talking NPCs, all of these would not be possible without AI. And the same is a lot of people don't realize the next generation of AI would not be possible without XR because for AI to really have

to get to the next level, it has to have real world understanding. And right now there aren't enough physical sensors for these AI to data to understand the real world. So a lot of its training for the 3D world is actually done in virtual simulations. So it's kind of like a matrix for AI, where we put them into these virtual worlds, and this is how robots are now training each other. This is how autonomous cars are training how to avoid accidents and to handle different edge cases in the world. We need that type of systems to be around. In fact, if you look at all the new tele-operated robots now,

all the humanoid robots, they're all trained using tele-op systems where a human is wearing a headset, doing something, and the cameras and the sensors are able to capture what the human's doing. And that information then informs the robot how to do something and to be able to repeat it at a professional level. So that type of tool is mandatory for us to get to that next level of AI advancement. Now, the other thing that a lot of people maybe don't want to understand or maybe are afraid to see is there's going to be massive job displacements that will come from the maturity of AI, starting first in the white-collar world. And then later on, as these human or robots become more and more pervasive, even the blue-collar service work will start to be displaced as well.

And what would these people need to do to retrain themselves? I think XR will play a significant role in terms of being able to retrain the humans into new skill sets that allow them to be productive in society. The other thing is there's also going to be a lot of mental stress that will happen from job displacement because so many of us right now, so rate our personal identity and link it to our jobs. Sometimes when you meet somebody, the first thing you say is, oh, you know, I'm Russian and I work at so-and-so company, or I, you know, I do so-and-so work or what's my so-and-so title, right? So we're so identified with our jobs as we are. And if, if sometime in the near future, a lot of people, some large portion,

10, 20, 30, 40% of the population is no longer working in these jobs that they used to be in, how will they have a sense of identity, a sense of belonging, or a sense of value to society? That's where immersive technology can play a major role, particularly if we can actually establish that metaverse infrastructure, where we can allow the global trade of service and labor and cognitive knowledge around the world, if we can allow for people to be able to do virtual cultural learning, virtual services child care whatever right where you can create virtual worlds and then use that as a way to to have an outlet for your energy or virtual art or virtual music or you can provide a service a virtual service to a physical person or company somewhere else in the world through a 3D cross-border

platform. So that's why I'm a big advocate of us starting to think ahead before all of these displacements happen to start building that infrastructure that we need to enable essentially a parallel tier of digital and virtual commerce to happen. And this also allows for easy and natural redistribution of wealth between the have-nots in the world and the rich and the developing countries in the world. Because at the end of the day, it doesn't really matter where you sit or who your parents were. You can provide a service that's valuable somewhere else and somebody else is wanting to pay you and compensate you for that service. So this allows that global workforce to be much more accessible than it is even today. I think you bring up so many good points, Salvin. I mean, one thing I think is not maybe spoken about as much is how the immersive technologies

and AI is actually going to be great to upskill. We hear a lot about potential job losses, but we don't really hear much about what we could do to upskill using these technologies and how it would be so much easier to be able to seek knowledge, learn, and be able to acquire new skills. The other point that you also bring up about equalizing, which is I think the next area that I was going to go into is how do we ensure that this is actually something that is not only limited to certain geographies or certain economic stratas, but is really actually equitable across the globe so that people have access to this technology and, you know, the communities that need it most in terms of

upskilling are able to access it. The other thing I think is also critical is while we are talking a lot about the fact that there's all of these amazing uses of AI, but there's certainly some guardrails or, like you said, how do we ensure that people are thinking about how we can encourage this adoption, but within these guardrails to allow for this to be a positive sort of like a an uber level you know across the globe are there organizations are there policies that are being put in place there are definitely a number of industry organizations that are trying to create more standardizations across entries and across vendors there are some regulations from policymakers that are probably more domestic or national level. I see more international level AI regulation, but very little international level XR or metaverse related policy discussions. So I think a lot of people have been, a lot of the ideas and thoughts around immersive tech was actually drowned out a little bit the last few years with the rapid

rise of AI and everybody's attention has now focused on AI. I think what a lot of people don't realize is that the issues that AI will bring to the world in terms of the job loss, the lack of progress, maybe even the lack of understanding alignment, the interpretability can be solved when we have a proper global metaverse system. So one of the things with AI that people are afraid of is, you know, AI is going to go rogue, it's going to kill us all, and it's going to become evil and domineering and etc. And the reality is actually, we could probably take these early AI models, put them into these virtual worlds and see how do they behave when they live

in a world of a million virtual beings. Do they act aggressively or are they more benign? Are they more compassionate? And to run it for a million years or a billion years in virtual time so that we don't have these issues where people say, oh, well, it's just pretending that it's nice until you give it all the power, then it's going to become evil. And you hear a lot of fear out there about where technology could go. But I know a lot of ways that I feel like we're anthropomorphizing it based on what we've seen humans doing, but there's really no evidence and no research that shows that more advanced AIs is going to be somehow more malicious than less advanced AI. In fact, I think that the bigger danger that we need to be thinking about is really more around the misuse by humans.

Oh, that there is 100% certainty that there are bad actor humans in the world. And when they get access to special capabilities, they will misuse it. And we were talking about cyber security earlier and before this chat. And if you're dealing with hackers and ransomware makers and terrorists, they're going to be using this technology to their advantage

because it is a very asymmetric tool in the sense of the person who has it, if you know this hole first, if you see a vulnerability before the people, you're able to take advantage. And that asymmetric capabilities

allows a very small group of people to have a very large influence on the system, whether if you want to hack into a system or you want to hack, or you want to turn off the power grid, et cetera, right? So there's a of things that that same type of asymmetric nature will will will be possible in the ai centric world when we have very advanced ais available especially as they

become more and more easily accessible as as well as easily processed, right? Because now a lot of them are being optimized, so they don't necessarily need to run on giant data centers. You know, at least with data centers, you can sort of monitor what's happening. But when you start able to run it at the edge, those kinds of security possibilities kind of go away. You know, so there's definitely a lot of potential downsides

and the greater need for the guardrails that we had talked about. possibilities kind of go away. So there's definitely a lot of potential downsides and the greater need for the guardrails that we had talked about, because the data misuse is really where we need to spend more time focusing on. Absolutely. What would your advice be to young technologists, folks who are in computer science, who want to get into this field? What would be good areas of research or work that they should be focusing on to actually make a significant impact? That's it. The pace of change that we're seeing,

essentially, if you're not already an AI scientist, it's probably too late for you to start to get into that field and to make an impact, right? Because there are already hundreds of labs and probably hundreds of thousands of people who are working in this field trying to advance it. And at the pace of change that we're seeing, within the next one to five years, there's a very high probability that we will get to AGI and then likely within months or years of that, we will get to ASI. So I actually am less, I guess I would be less focused in telling young people who are just starting out their career or starting out their studies in computer science to

say, okay, now you should go be an AI scientist or a researcher. I would say, look at what the AI systems have today and find an area where you are passionate about, whether it's for healthcare or whether it's for education or whether it's for better food or helping the climate or whatever. I think there's a lot of, the value will come more from the allocation of this technology than it will come from the development of this technology by the time that they start to enter the workforce. And honestly, I feel like a lot of the people who are in these industries today, they don't have enough technical basis to properly apply what's out there. So this may actually give young people a greater voice and a greater influence in whatever industry that they go into

by applying what's coming out of the labs into the real world. Because what we actually find right now, even with what we see, the advances in capabilities that we see in things like the GPT models or the anthropic models, their actual use case in the real world is still relatively limited.

And it's something that will not be the case within two or three years where almost every senior leader I'm talking to is trying to find ways to apply it. So if you can think about how you can apply it to solve real world problems, whether it is cognitive problems or data analysis problems or decision making problems or robotic problems to solve physical issues, I think all of those would be amazing use cases. to do as much of what you can in the weaponization side. I know there are people in the defense industry that are very heavily focused and trying to apply this technology to creating better weapons and better ways to harm your opponents.

But I think we need to be careful about the type of signals we are sending to these AIs and how we're training them. Because the AI is a reflection of what you tell it to do. And if you're telling it to find the best way to kill somebody versus finding the best way to solve the climate problem or the food problem or the health problem, it will do its best to do those things.

And I would rather we have more of its capabilities and energies and more of our creativity and innovation put into the positive use cases. Very well said, Alvin. I think that's very encouraging and also such a positive direction in which to sort of motivate the youngsters of the world to start thinking about the applications of this technology, to embrace it, to learn more about it, and apply it to the problems that they're most passionate about. Thank you so much for your thought-provoking work and for taking the time to speak with us at ACM ByteCast. Well, thank you for inviting me.

I enjoyed our chat and happy to chat again if you ever would like to do that. Absolutely. Thank you, Alvin. ACM ByteCast is a production of the Association for Computing Machinery's Practitioners Board. To learn more about ACM and its activities, visit acm.org.