Name: Brainstorm AI Singapore 2024: How AI Is building the cities of the future
Uploaded: 2024-07-30
Duration: 18 min 38 s
Description: Cha-Ly KOH, Founder and CEO, Urbanmetry Shaun KOO, CEO and Co-founder, H3 Zoom.AI Joe XIA, CEO, Jidu Moderator: Nicholas GORDON, Editor, Asia, FORTUNE; Co-chair, Fortune Brainstorm AI Singapore

Cha-Ly KOH, Founder and CEO, Urbanmetry Shaun KOO, CEO and Co-founder, H3 Zoom.AI Joe XIA, CEO, Jidu Moderator: Nicholas GORDON, Editor, Asia, FORTUNE; Co-chair, Fortune Brainstorm AI Singapore

Brainstorm AI Singapore 2024: How AI Is building the cities of the future

Conferences

So Charlie first question to you. Um you know, a lot of places talked about being smart cities and the benefits from them are sometimes a little hard to recognize. Um there have been some pretty high profile failures as A I really developed as we get this new technology is the idea of the smart city now maybe coming into fruition. Thank you so much for having me today. And I think I'm really excited about the potential of A I primarily because it levels up the playing field. So for the longest time, smart city was pitched as this innovative, very, very effective way of governing cities, but it was afforded only by developed nations or developed cities, right? So the four processes to achieve a smart city, data collection, data analytics, decision making and action. For the longest time, the folks who can afford data collection was really developed cities like Singapore, many parts of China, Korea today, data collection can be done at a fraction of the cost using A I for developing cities, right? So that makes it bringing it forward in terms of leveling up the playing field for these cities. But I think to, I also wanted to curb some of enthusiasm, right? So like if you think about data collection that levels it up data analytics, great. You can now use some of the A I to do that. But let's talk about a little bit in the data decision making and action is that purely able to be done by A I in this part of the region. I think we are still quite far away from that because fundamentally cities are political and there are political parts of the decision making that unfortunately, we don't trust A I or I don't foresee that we trust A I to do that completely for us. At this point. I do want to get back to this point about politics and trust. But before I do, I want to bring in Sean, you know, as someone who is doing this data collection, could you tell us a bit more about A three zoom A I and what you're trying to do? Right? So at Atrium Zoom A I, um we see ourselves as the pent of the build environment. So we provide intelligent insights and actionable outcomes for property developers, fortune 500 property management companies and engineering consultants on the entire building condition assessments, essentially maximizing the life cycle value of these assets, right? So, so that's essentially like what we do, we work with a myriad of data sources, drones off the shelf cameras and we just ingest all these data sources to provide these outcomes in a very intuitive form. Um And I want to bring in Joe now, you know, autonomous driving is another one of these terms has been talked about for a long time. You're with G do, could you explain to us kind of what exactly is G do and how it relates to Baidu? It's A I initiatives with Ernie and the Apollo Robotaxis. Where does G do fit in this network? Uh Yeah, thanks for having me. Uh I think uh uh J uh first of all J is uh we're kind of gonna start up company. We're only three years. Yeah. Um So when we found it, um we're aiming to actually developing, we call uh smart EV, it's not ev it's a smart EV, right? Um So we call our product A Robocop. Um So, um so we have a joint uh shareholders, one is back to another, the GD. Um The goal for them to come in together is, you know, for the past 10 years that have, you know, that has been like invested over 100 billion um IMB on A I technology. Um So, so it's just the right time right now for uh for us to bring those A I technology, you used to be wrong on the cloud now, we can bring that to the cloud. So, so that's why it brings us here. Um And, and, and I think it's the right time because the computing power on the car side and both on the computing side on the car side are, are right now, is big enough to supporting those air to running. For example, um A 04 used to be like, you know, a very heavy on computing on GP US and you don't, we don't have a vehicle level um chip set in order to run that on consumer. Um you know, vehicle uh not until 2000, 2023 we have or X come from, you know, came from a media. So that's why we're doing that is uh uh uh we, we, we, we, we develop a car and we in great a be A I technology like board and uh you know, TB like, you know, A I um program and uh A AL four technology, but not exactly L four. But uh how are we gonna uh firstly, um you know, kind of downgrade the L four and having customers to be able to use a high level, you know, kind of atomic driving in the city. So that's the, the purpose of the goal of, of, of ju Yeah. All right. I want to stay on this politics point now, maybe first question to you, Charlie. Um You know, I, I've read some of your commentary on this and you, and you talk about things like housing, housing affordability, big political topic. I saw your linkedin Post about diesel subsidies in Malaysia. Another big political topic. How are you, how do you see data analytics and A I and data collection interacting with the political discussions in these cities, particularly in Asian cities. No, I think for the longest time because data was so hard to collect and things become a little bit more opaque. It was very hard for a data driven decision to enter into the conversation. I'm I'm saying enter the conversation because things are moving very quickly that by the time you collect this information, it's too late, right? But today, things are so real time that the conversations and the disparity between some decision making and the data that's available becomes a little bit hard to reconcile. So I think the availability of data gives a lot of transparency in terms of what are, what are people really being motivated by, by doing certain decisions, right? So the diesel point that I made on linkedin was that we are consciously thinking about how we are building a city very far away and that in itself is a conscious decision on how much we're spending to transport our logistics. So while we are solving it with ev while we're solving it with efficiency of A I, shouldn't. We also use this to plan in the future and reduce that today. So those are the part of using the data collection using A I will help us bring that point forward to be part of the conversation. Not really to dictate it, but at least participate in, there's also a risk that, you know, people can interpret the data, how they wish in ways that support certain conclusions. I mean, what's the risk of people then misinterpreting that data and thus worsening the political debate over a certain issue, I think uh that's always the case with any kind of data and however you um you collect them, right? And I think a lot of times people are really afraid of using data A I to do analytics because of biases, right? But I think it is even worse if you don't have the data available for people to not be able to actually drill down to to to even question the data. Data transparency is very, very important and the ability to allow general folks actually criticize that data, I think is step one. I want to go back to Sean now, you know, when, when drones were kind of the next big thing in tech, you know, they were cheap, they're vous they were everywhere and then we all got distracted by something else. Um You know, the two, you know, what else can drones help us do in terms of urban planning? And then how do you manage people's perception of those drones? They may worry about them being invasive, they may be worrying about personal privacy. So how do you chart that balance in terms of taking advantage of what drones can do? But also making sure urban populations aren't scared by fleets of drones like data on them, right? So, so, so that's actually a very good question. So, so obviously, um I think there's a lot of fixation on drones as being like the next big thing, but obviously, you know, through sort of conversations with partners, customers as well as industry um um stakeholders um at the end of the day, um what the the data, right that is being collected by drones, be it whether it's for construction monitoring, uh for asset inspection, for precision agriculture. Um what the industry and the customers really do care about are the insights, right? What are um actionable outcomes and what can um A I algorithms um that have been developed, you know, how, how can they actually interpret some of these unstructured data? Right? So, so I think the the the the perception definitely has changed over time. Um And, you know, from, you know, from my perspective, uh you know, being in, in, in, in the business for like close to 10 years now, um the sort of evolution of acquisition has also been extended to ground robotics, wall robotics. Um because, you know, the end client, what they really value and care about are outcomes that would drive the decision making processes, right? So I think in terms of governance around the data collection um and security aspect, I mean, there are, you know, existing guard rails um that um that are in place, right? So obviously there are algorithms out there that could, you know, mask uh faces windows, um you know, number plates, et cetera to sort of appease um and uh appease some of those like regulatory requirements. On the other hand, you know, there are um you know IO standards um that, you know, can be built right within um the uh the system level, right, in terms of data, encryption, sovereignty, etcetera, so that there are definitely like best practices out there, right? So, I mean, in short, I guess to drive home, the point is um you know, the the the the end the end outcome of being able to interpret the unstructured data to drive the business decisions are the true sort of a testament of what the industry and the clients are actually be looking over for. All right, one more politics question, then we'll move back to safer territory. Um You know, so Apollo's Robotaxis investors have been very excited about those, but I know that led to some controversy in Wuhan from people worried about how they will affect jobs. And you know, how do you, how should cities manage that transition as more stuff gets automated? How do they manage the human labor that, that gets affected from that new technology? Yeah. II I think uh uh just, just quickly answer the question. Um It's still far, it's still, you know, too early to think about that, right? Because uh um you know, uh Baidu has like started developing autonomous driving since 2013, um like 11 years ago. So the government under actually the supervised supervision of the government. Right. Um So, so, you know, started from a single vehicle, a few vehicles on a single road, you know, all the way to now, you know, uh just to be clear, not until last year. Um you know, um um um 40 L four robot taxi is not allowed in China to operate in the government just open that up last year. And, and, and then really with a small scale, not a large scale, talk about Wuhan Wuhan has over like 40,000, you know, taxi and, and, and the L four road taxi right now is only like 405 100. It's very small amount of road taxi. I think it's still under trial, the government, it's not, it's not there. It's not like the next day they're gonna be like thousands of hundreds of thousands of, you know, p in the road, right. It's, it's, it's still, you know, too early to, to worry about that. Right. I mean, still it's still a trial and, and, and, and most importantly, I think every time there's an evolution of technology it's going to open another window up, right? And you think about it years ago when we talk about, you know, um um you know, those robots on car manufacturing lines right. People thought like we're going to lose, lose our jobs, right? But turns out there's more jobs that are created because those robots, when they, those robots come in and, and, and it's actually increased the efficiency of the manufacture. And in that case, actually it makes the the sales of the car volume even bigger and it creates module, right? So, so I don't think we have to really worry about that at this stage. So automation will create more labor, I think ended up will create more, more jobs, you know, similarly on this labor point, I mean, um by, by replacing, you know, data collections, like people, you know, guys on the street with clickers, wiring the traffic with sensors. Um Yes, I mean, you will be, you'll be replacing jobs would normally go to people. How do you see that transition again? Also, especially in most of Asia where, you know, labor is probably still cheaper than capital, you know, how do you kind of get balance between labor and technology? So I think most of ASEAN is still thinking of the current situation, but A I is going to be built for the next 50 years, it's not going to be for today. So if you are thinking of your population as the next, of course, everybody needs a job. But is it for the job today or the job for the next 30 years? Do you want to be a guy standing there doing the quicker or the guy who's actually designing the A I system, who's counting the cars. So governments are actually in the position right now to make a call around the transitioning towards the higher value jobs in which you can actually create the value in the next 2030 years and not having to worry about stagnation, right? So stagnation is something that we should worry about today and taking the the step of actually using A I as augmented uh you know, help for you to, to do that is going to be a good system. I think um Sean, I do want to bring in the sustainability point, you know, making cities more climate friendly. Obviously, we're in a region of the world that's going to be affected by climate change. How will these technologies make cities more sustainable? Right. So, so obviously, um basically, the build environment is kind of like the bedrock of urban city centers. Um Obviously, you know, if you're using like robotics and you know, drone platforms, et cetera, um there are a lot of direct benefits um in terms of replacement of having to like ferry workers on site specifically in Southeast Asia, right? Because we still rely on a lot of low skilled foreign workers that are, you know, you know, essentially building the buildings and the city centers like from ground up, right? So obviously, in terms of carbon emissions, there are like indirect impacts on transportation that's one in terms of the data aspect of things by being able to plug into various data sources within building management systems. HV AC. There there is a potential to also reduce the carbon emissions from buildings, right? So 30% of carbon emissions today, you know, they come from buildings, right? So if we can reduce sort of like the footprint, in terms of how buildings function today through some of these insights, then it would, you know, empower architects and designers to design more sustainable buildings and essence, right? So, so that's how we, you know, how, how I see, you know, sustainability, I want to squeeze in a couple more quick questions for the time. We've got Joe, you were also previously before g do you were the co founder of Mobike, you know, bike sharing app, bi bike sharing service. What's the lesson from that experience that you think applies in terms of data and urban planning? Yeah, we actually um so, so people are just simply think Mobike is um bike sharing and we just dump bicycle on the road without thinking, right? And actually not, we're actually utilizing a lot of data. Um So um we actually work with a lot of transportation um research organization in China. So what we're doing is we're actually using the bicycle riding hotspot um like map called map and, and combine that with bus and taxi and those like, you know, ride hiring, you know, see how these, those three different types of transportation is gonna make up the whole transportation system for the, for the city. So we actually help the government to re replanning, replan their bus station because of the bike sharing, right? Because sometimes the last mile can be, can be covered by back sharing. And we actually use those data. Well, well, it's anonymous data, right? Um you know, work with, with government like combine those three system like data from different systems and and and use that for urban, you know, kind of planning, you know, to make people's like life in the city more easier. Yeah, super last word Charlie, how much is this like the game sim? So like the games in city, when you interact with the city, there will be an effect, right? I think the most of the time people forget that the people protest too. So I think if you remove, if you increase taxes and you start monitoring everybody, there is a danger as well of pushing them too far, they start going on to the streets, right? So I think as much as we love A I and data and data collection, we have to be wary that being in a city and being in a urban setting, it is a political thing and decision makers can use it to assist in collection of data and the data analytics. But beyond that, I think we do have to be worried about it. OK. Last time we got, please join me in taking all of our speakers.