AIBP ASEAN B2B Growth
AIBP ASEAN B2B Growth
World Bank: Ozzeir Khan on AI as Southeast Asia’s Growth Engine
In this episode, we are joined by Ozzeir Khan, Director and CIO | Operations and Country Solutions, The World Bank , as he discusses the transformative role of AI and technology in Southeast Asia’s development journey. Reflecting on his leadership at the Asian Development Bank, Ozzeir discuss's the region’s unique ambition, its communal approach to technology, and critical considerations for governing AI as a potential public good.
Tune in as Ozzeir highlights how ethical frameworks and responsible innovation can shape a more inclusive and impactful technological future for Southeast Asia.
World Bank: The World Bank is an international financial institution dedicated to reducing poverty and promoting sustainable development worldwide. Established in 1944 and headquartered in Washington, D.C., it comprises 189 member countries. The World Bank provides a wide array of financial products, technical assistance, and policy advice to developing nations, aiming to improve economic prospects and quality of life.
Asian Development Bank (ADB): The Asian Development Bank (ADB) is a regional development bank established in 1966, headquartered in Manila, Philippines. Owned by 69 members—49 from the region and 20 from other parts of the world—ADB is committed to achieving a prosperous, inclusive, resilient, and sustainable Asia and the Pacific. It provides loans, technical assistance, grants, and equity investments to promote social and economic development in the region.
Throughout history, major societal leaps have been driven by technological revolutions. In this episode, we explore how Southeast Asia is embracing technology, not just as a tool for growth, but as a powerful force to uplift communities.
Ozzeir Khan:Consider myself to be a development technologists, and I've seen the sort of the fruits of how tech can drive development. If you look at the history of the world, actually, most of the big leaps in societies has been on the back of some technology revolution.
Unknown:Southeast Southeast Asia stands out with its bold technological ambition, pushing beyond digital transformation towards smarter, more livable communities, emphasizing shared societal growth.
Ozzeir Khan:Twins is all about how we can make it not just smart, but make it livable in the way the future is being shaped. The ambition I see in Southeast Asia much higher than other countries, the trust in tech much, much bigger in in Southeast Asia. And then one other unique element that I find in Southeast Asia, different than, let's say, other other continents, especially the West, is the interest in communal, good versus personal.
Unknown:With this ambition comes responsibility as AI reshapes industries and communities, critical questions emerge about ethical use and responsible governance, highlighting a shift towards understanding AI as a public
Ozzeir Khan:So we're just entering this AI age, so to good. speak, and to me, it's it's closer to the start of sort of the industrial era than it is the digital it is going to bring in a lot of fundamental changes. And there are few big questions that we'll be answering in the coming decade or so, and one of those, big one is what needs to be governed as a public good.
Unknown:AI's rise in Southeast Asia brings important considerations to the forefront, especially regarding ethical governance and the potential treatment of AI as a public good, how the region approaches these considerations will impact both economic and social progress. Please enjoy this episode from the AIBP ASEAN B2B growth podcast.
AIBP:The AIBP ASEAN B2B growth podcast is a series of fireside chats with business leaders in Southeast Asia focused on growth in the region. Topics discussed and include business strategy, sales and marketing, enterprise, technology and innovation.
YY - AIBP:Hello and welcome to the AIBP ASEAN growth podcast, where we sit down with individuals responsible for driving growth of their businesses here in Southeast Asia. My name is YY, and today with half with us a very special guest, Mr. Ozzeir Khan in his very fresh role at the World Bank. Well, Ozzeir, your career has spent multiple roles at United Nations, global financial institutions and Asian Development Bank and now at World Bank, share with us a little bit more about yourself and how your diverse experience has shaped your perspective on technology in a development across Southeast Asia.
Ozzeir Khan:Thank you, YY, and very nice to be here with you. A little bit about myself. So my background, I did my college degrees in computer science and sort of global finance. So when I came out of college, I was like, Okay, should I do computer science, or should I do banking? And I would like scratching my head, and then I had a lot of interest in globalization, because that was what was going on at that time, in early 90s. Unlike today, what you see the globalization with the big mantra at that time, and technology was driving a lot of that. So I got involved from very beginning in this intersection between finance and tech, initially in private sector, in management consulting, and then moved into United Nations in New York. Most of my career has been in what I call development tech, and I call it development tech, rather than Tech for Good, because here we have specific objective that we drive in United Nations, work for peacekeeping operations in Africa, and development operations in some of the low income countries. And then recently, for the last seven years, I was with the Asian Development Bank, leading digital, including AI efforts, and now very fresh, starting global operations in tech for World Bank. So very happy to be here at at World Bank, but my experience over the years, primarily, I consider myself to be a development technologist, and I've seen the sort of the fruits of how tech can drive development. If you look at the history of the world, actually, most of the big leaps in societies has been on the back of some technology revolution. So it's been great to be part of that, and we'll continue to do that in my new role here at World Bank.
YY - AIBP:Thank you very much. I see that you found the perfect balance right between finance and technology, and now you are looking into it at a World Bank stage. And you mentioned that when you first started, it was globalization. The landscape has changed quite a bit by thing development, technologies. It's even more important in this current landscape. But if I look at Southeast Asia, i Ho she's in a very unique position. How do you think it differs within Southeast Asia compared to perhaps the other regions that you have had experience at?
Ozzeir Khan:Yes, so I work globally, in pretty much all continents, and what I find about what I find different about Southeast Asia, especially when it comes to tech, is the ambition, the ambition on digital, the ambition on AI, the ambition on technology, on how it can change. And it's all across Southeast Asia. It's not necessarily just Singapore. A lot of people think that Singapore is the center of all sort of tech, but when you look at Indonesia, Jakarta, you look at Thailand, when you look at even Philippines, the gaming industry in Manila, you see thatnot only there's interest, but A lot of risk taking and investments in various components that make up some of the major infrastructure reforms in these countries. For example, in Indonesia, if you look at the new capital being developed, the new capital in Indonesia is all about digital twins. It's all about how we can make it not just smart, but make it livable in the way the future is being shaped. To the ambition I see in Southeast Asia, much higher than other countries, the trust in tech much, much bigger in Southeast Asia. And then one other unique element that I find in Southeast Asia, different than, let's say, other other continents, especially the West, is the interest in communal, good versus personal. So here we see a lot of sharing of tech, a lot of sharing of data for for good, for betterment of the country, or for the town or city. So a lot more on the ambitious side here in Southeast Asia.
YY - AIBP:Thank you very much. I believe we also see the same thing right when we are tracking like what kind of technologies that the ASEAN enterprises or organizations are looking to apply, there is a lot of interest in, say, AI, but I think when it comes with ambition, when you're looking at Investing in this technologies like AI, there's this very important measure, right? You need to be able to measure the successes, or perhaps failures, of your innovation efforts, right? Can you share with us in your experience? How do you measure successes, and what does maybe the idea of innovation mean to you.
Ozzeir Khan:So the success when it comes to development work, when it comes to moving sort of humanity forward, and when it comes to saving sort of at the planet, there are a few sort of key aspects that we look at. It's not just overall impact and impact you can sort of measure in terms of how many people you have moved from, let's say lower income levels to middle income levels, right? So that's sort of the impact metric that you look at. But on top of that, you kind of focus a little bit lower and you. Look, you see, what are the outcomes, not just the broader impact, but the outcomes that you need to focus on to achieve that bigger impact, and the outcomes you need to have cities where you can breathe air that is clean. You need to have cities where the traffic can be managed. You need to have a city where the crime is at a level where people can can invest so the outcomes matter. And then lastly, success is also about outputs, what specific things you're trying to produce. So for example, if if you want to create a certain pricing AI model, then that model needs to be produced managed as an output. So these three components together can form a very complicated equation that needs to be resolved for success when it comes to development and with sustainable development goals, it there's so many different aspects that need to be considered. So this equation becomes complex, but resolvable more and more especially now with with AI
YY - AIBP:Understand. I remember that you were sharing with us about your digital innovation sandbox program, and you mentioned about your proof of value test, right? Because, beyond just moonshot projects, sometimes you need to look at the how realistic it is to implement the project. Can you keep some specific examples of what you have done and maybe some of the programs that you have run of how you decide what kind of measures that you take in in determining the outcomes that you've just mentioned?
Ozzeir Khan:Yes, the sandbox program that we ran at Asian Development Bank and is still in production. It is unlike some of the others in the industries, which focus more on experimentation or proof of concepts or understanding a success of a certain technology, it really focuses on actual testing in a real environment with real citizens. For example, in in the sandbox, we did a project in Papua New Guinea where in in parts of the country, people had no IDs, and they lived in places where there was no internet and no connectivity, and they had no access to banking. So financial inclusion issues, identity problems, but these folks in very far off places, did fishing. They had commerce going on, but there was no way to get them into the financial ecosystem, to provide loans, to provide other sort of financial instruments for their growth. So how do you resolve this with with with technology? Of course, this is not something that you can pick up the phone and call a major player and buy something, right? So you have to create something, and the private sector will never focus on where commercial viability is low, because there are only million people with this problem, it doesn't scale. So the private sector will not get involved. So sandbox program in development tech becomes very important, where institutions like United Nations or World Bank or development institutions like ADB go in there and create solutions which would be relevant. So the sandbox created a solution which was offline, was able to put kiosks in remote areas and then have a mechanism with the local government to have the sovereign data in the central bank's data center, owned by the government. And the trust aspect with the local financial system, in terms of banking was, was, was developed. But the key there was to not just prove the concept, but to implement that concept in a certain village, a certain town. And that's sort of the key element in terms of sort of testing it out. The way I describe it in some of the other places is it's like almost when you do a drug approval, you have to do human sampling, you have to take it out there before you get national approvals. And the sandbuff programs that we run in development sector do go into actual usage in countries.
YY - AIBP:And I if I could just highlight one of, like, the the projects that was done in the sandbox program. I believe one of it was about financial inclusion. And when you look at like how perhaps the smaller MSMEs, perhaps do not have, say, financial data or credit rating. You were able to use alternative data. Can you share a little bit more about what happened in those projects? Because I believe now this is a problem in many parts of Southeast Asia, and I really like the fact that you know when you face a problem, you don't you look at it from an alternative way in order to address the issue,
Ozzeir Khan:Yeah, I mean the small and medium enterprises, and I would even say micro enterprises, where the issue happens in Philippines, you have these size, size stores, which are all over the place. And how do you provide financing to them? A lot of that is supplier based financing and based on stocks that they own or they have to replenish when they are sold. So coming up with solutions which are very local and data driven, and where the intelligence piece is brought in, because what do banks do? Typical banks for large customers, they have humans understanding the financial statements, understanding the business and giving a loan. So now with AI, we have this natural ability to now have AI do that work, and for AI to do that work, it can get data from all kinds of sources, but to localize it to the micro enterprises, to actually put sensors right there on their stockpiles, right at the point of sale, so that the financing is fit for purpose. It's not that your blanket approving X number of dollars per store, but it's very, very focused on that particular micro enterprise. Previously, a lot of the issues that we faced in this industry was the small and medium sized enterprises were lumped into segments, and then this segment gets this amount of money, and it wasn't. It was very difficult to do it for that particular small store because of the volume, because of the size.
YY - AIBP:Thank you so much. You know, you've mentioned about, like, Sari-sari stores in the Philippines, and I think in Indonesia, there's war rooms, right? And I believe that in Southeast Asia, the region presents a very diverse landscape, and it is the same for regulatory approaches to AI and data, right? So let's say, in your experience, if a business has multiple countries in Southeast Asia that are navigating this data and AI regulations. How can they best manage these differences while still being able to scale their digital solutions across multiple markets in the region?
Ozzeir Khan:First of all, I think the understanding of AI versus digital and in many cases, what happens is that AI is looked upon as a bit of a sprinkle on top of digital, and which technically is, is not true. And because we're running a lot of AI projects, and the way we look at this in development world is that, if you look at history, how far analog is from digital, we see the same gap between AI and digital. See, we normally don't think of that gap, but the problem we were not able to resolve with digital is because of the limitations of digital. The problem we were not able to resolve before was because of the limitation of living in analog world. So as we enter the AI world, one of the things we have to do, especially the problem that you mentioned, which is multiple countries, different cities, different cities, different cultures, different ways doing things. If you apply digital, it doesn't is tough, because digital assumes that you have one central approach, one Airbnb, one Uber, one kayak, one platform, one this one that. Versus AI, which you can have 50 different AIs for each one of those areas, tailor made, focused and then managed very differently than you would as a digital platform. So these fundamental changes as you move into AI are really opening up opportunities for the problem that that you mentioned, because you're right. I mean, what happens in Mindanao versus what happens in parts of Chiang Mai, if you put the same platform digital, it doesn't work. But the AI model, you can train it with the data from that region, and it's only for that region, and it can be done with the same AI workforce or lab, or whatever structure that you have, you can produce them now. So, yeah, huge, huge opportunity.
YY - AIBP:So in this case, like, what you've described as like AI is infinitely more scalable right across the different countries, and still like digital where perhaps it's a lot more limited by what each country presents. But let's say, you know, the models that the AI models that you've mentioned, is scalable, but we still see that the speed of AI adoption across Southeast Asia, it's varied, right? What do you think are some of the factors that are causing the differences in AI adoption?
Ozzeir Khan:So there are two things here, AI adoption and AI production. So the production side of AI also needs to be scaled for the adoption to happen, because they have to be tailor made for the situations that we just talked about. The approach of having blanket AI solutions and then expecting them to be scaled is tougher. So this issue of production of AI and consumption of AI goes together. So as we move forward, we see, and we're seeing a lot of activities now in terms of production of AI, which is tailored to the consumption areas in AI and especially in Southeast Asia, in different sectors, from transportation sector to energy sector. I was discussing with someone the other day that with AI, now you can have a pricing mechanism for electricity which is based on your ability to pay, rather than the whole village being offered the same pricing. So that allows you to have overall more sort of optimization for the utility company, which you can do with an AI platform you cannot do with a digital platform, but that requires the utility company to invest in AI infrastructure to manage
YY - AIBP:That is something that I believe the energy companies in Southeast Asia will be very interested in, right? I think when we speak with PLM, that's perhaps one of the areas that they are looking at AI to solve. Um, you know, if we talk about AI, and just now the example that you've mentioned in the ideal scenario, the AI tools in energy sector will allow perhaps a more equitable distribution of energy. But AI has also brought into the picture a lot of questions around responsibility, the gaps, whether the haves and have nots. How? How would that be impacted with AI moving at breakneck speed, right? You know, in your writing, you emphasize about the ethical guardrails. You just make sure that AI systems, you know, not only respect human values and rights, it's also something that create good for the community. Can you tell us a little bit more about maybe the frameworks that ASEAN organizations should adopt to just ensure that AI can benefit the population?
Ozzeir Khan:So we're just entering this AI age, so to speak. And to me, it's, it's closer to the start of sort of the industrial era than it is the digital so it is going to bring in a lot of fundamental changes. And there are a few big questions that we'll be answering in the coming decade or so. One of those big one is what needs to be governed as a public good. So for example, in countries you have, let's say, a defense which is government, you have some bridges and roads which are managed by the government and not private sector. AI is intelligence. So what part of intelligence needs to be managed as a public good versus part of a private sector? And I think this, this conversation is is beginning, and there would be aI public goods which will be managed more as a public sector activity, and that will bring in some of the big issues that we were discussing in terms of guardrails in China, for example, if you look at the data coming from all the automobiles, it goes to the central government, government and gives that data to all the AI companies to use way different than the Western countries on how it's being done with the private sector. So the guardrails in terms of what is needed are very well defined in in a lot of the frameworks which you see today, from ISO to NIST to others. But the adoption of those requires us to have governance, requires us to understand the difference between public AI versus private AI, and those are not yet clear in terms of governance. And this is why, when people have issues, it's not necessarily that we don't know what needs to be transparent, what needs to be fair. It's more about the governance around that, and that's an evolving area. And I'm a firm believer that there are parts of intelligence that needs to be a public good.
YY - AIBP:I like that. I think we were just having this discussion in Thailand at an advisory board meeting where we were discussing whether the government can take up the role of driving more AI for public good as a public good, right, and in furthering the AI adoption within the country itself. So that aligns really nicely. Well, you know, we've talked a lot about the opportunities, talked about your experience, also the fact that Southeast Asia is a very interesting space for you, because the you see the nations being more ambitious with technologies like AI. But if you want to leave our audience with one key message about the future of AI in ASEAN, what do you think that would be?
Ozzeir Khan:The biggest sort of opportunity that we see in AI is applying AI across multiple sectors. And when you do that, there are areas which are shareable between sectors, and we need to figure out how to do that even in private sector. So if there is different let's say you have Association of Transportation, Association of Energy, and the AI being created have this layer that they need to share together, even in private sector, not just the AI public goods, and that piece is going to bring in a lot of GDP growth. And in the digital world, we have figured that out. It took a long time. It took 20 years or so to figure out what we need to share between different industries in the AI, it is just starting, and those who will figure out first is going to have huge economic growth. So if you think about what we discussed with energy, if there's development there, and that can be shared with the automobile industry, then the sectors together, combined, are going to grow exponentially faster than alone. So this shareability aspect between the industry requires focus in private sector, and I would highly encourage especially in ASEAN, where there's so much interest, so much ambition, so much risk taking, so much capital is being put against that there needs to be a layer which needs to be shared in private sector.
YY - AIBP:Thank you very much, Ozzeir, for your insights. I like that you're advocating for transparency, for sharing. I believe the world, as you first started out saying, globalization was a key trend when you first joined the industry, perhaps when we talk about AI, globalization and sharing of data will help to drive AI to the next stage of development. Thank you very much, Ozzeir, we will miss you in Southeast Asia, but I believe perhaps you can do even more good in your new role in Washington, and we look forward to having you more often in our discussions in AI in Southeast Asia. Thank you.
Ozzeir Khan:Thank you.
YY - AIBP:Thank you very much. It was great, and I really like your your context. So perhaps what we can do is like, in a follow up, right? We will see whether we can have some more deep dive into some of the areas that you've talked about, right? Because the data governance, or AI governance part, or data sharing part, is something that we are exploring in this part of the world, and it's actually really difficult.
Ozzeir Khan:So it is difficult, and the success, I think, will come maybe in the public sector. First, you know what we were discussing, the AI public goods. You know you talk about DPGs. AIPGs is the future in my mind. So if we do, you know, get people around the table on AIPG, which is, which is, which is, I think if you take few years forward, that's going to be a very big word in public sector.
YY - AIBP:We look forward to it. I was going to say we will want to have you in our conferences in Southeast Asia, but Washington is going to be a much, much longer flight than Philippines.
Ozzeir Khan:But let me tell you something, Asia is also part of my, you know, mandate. So yeah, so I'll be around,
YY - AIBP:Yes please. And when you have a travel plan to Asia, let us know, and we'll try to see whether we can do something that aligns with your travel plans in this part of the world.
Ozzeir Khan:Yeah, will do. Thank you so much.
YY - AIBP:Thank you very much. Ozzeir, you take care and all the best. See you. Bye, bye.
Unknown:We hope you've enjoyed the episode. For more information about business growth in the ASEAN region, please visit our website@www.aibp.sg. you.