260225 0840 Opening keynote Understanding the future as it unfolds the power of megatrends

It's now my great pleasure to introduce our first keynote speaker for today's conference, Dr. Mark Esposito. Mark brings a rare combination of academic depth, policy insight, and practical engagement with the techniques that are reshaping our world. He's a public policy scholar and a social scientist whose work sits at the intersection of technology, economics, and governance. He's affiliated with Harvard's Berkman Klein Center for the Internet and Society, the Center for International Development at Harvard's Kennedy School, and the Institute for Quantitative Social Science. He also leads policy clinics on technology governance around the world, helping to translate technological challenge into practical policy frameworks. Alongside his policy work, Mark is deeply engaged in business and innovation. He is professor at Hult International Business School and an adjunct professor at Georgetown University, He's actually co-founded himself several AI ventures, including Nexus Frontier Tech, the AI Native Foundation, and Chart Think Tank. He also serves as chief economist of a Silicon-based AI lab, giving him a front row view of how these technologies are evolving in real time. He's a prolific author with more than 150 publications and 13 books to his name. His work constantly explores how societies can prepare for and shape the future, rather than simply react to it. For a Solutions Summit focused on the future of universities, knowledge, and national capability, his perspectives could not be more timely. Mark, we're delighted to have you with us. Please join me in welcoming Dr. Mark Esposito to the stage.

Good morning. I think it's working. Thank you for having me. I commend you for having spent some time in remembering your colleague. I found it to be a beautiful contribution and a great way to remember the limited amount of time we spend on this earth. First of all, thank you for hosting me and giving me refuge in Australia. I escaped Boston before one of the most historic snowstorms that remind me of my native Canada. I thought that I would run away from those kind of events, but Climate volatility is upon us. Today I'd like to spend some time in walking through a set of conversations that are designed to inspire many ongoing dialogues at your universities. It's more about coming from some of the work that I've been pioneering for a few years now, and it's now being encapsulated by the concept of triple transition. As a reference, I work at Harvard. I've been part of the Harvard University for 15 years, and a large part of my work has been in the intersection between policy and business. And today, as a natural evolution of my work, I've been integrating more of the governance part, considering the incredible transformation we are witnessing globally. See if it works. The clicker doesn't like me too much. And I am pressing— yes, the green button. All right. So you might be asking yourself the question, when we're talking about triple transitions, what are we talking about? It is a transition that is not sequential, and it's not necessarily happening as a normal sense of linear extrapolation from one phase of history to the other one. It is a transition that is to some degree has historical legacy, but it's concurrent. We still have country today that are largely embedded into traditional industrial capacity by reimagining their manufacturing. We have country that are phasing out some of the first maturity on digital infrastructure and country they're equally embracing the shift from the private to the public sector by opening digital public infrastructure. As agendas for their 2030, 2040 manifestos. But the most important part of the transformation is moving into what we call the cognitive economy. It's really the economy where it's mostly codified, largely dominant by the level of technology that we today have, which is to some degree consequential to the great changes that we did in the '80s all the way down to the beginning of the millennial in having given access to billions of people with technology, but it doesn't necessarily follow a linear evolution. It does become a system on its own. So if we think about what the 21st century is offering as an opportunity is rethinking that the concept of growth is no longer coming from traditional boundaries or traditional factors of production like labor, capital, and, and land. Is fundamentally coming through algorithmic capacity. Thank you so much. Now, I guess you perceive that we are currently in a geopolitical turmoil. I am probably experiencing as a Canadian in the US in the epicenter of what I consider to be an unsettling sense of abnormal that we no longer have the ability to to come to terms with, at least not in the short term. But we also true that while we might be distracted and captured by the headlines, we are realizing that the shift has already occurred. We start noticing that information is weaponizing the way country are basically positioned themselves. We find that if before we were looking at traditional forms of competition, today whoever owns the code, owns the game and actually defines the game in the first place. Now the first part of my conversation with you is really to set up a bit of a global stage where no matter where we are, these phenomena are currently happening. In the second part of the conversation, I'll try to narrow it down to Australia, but I don't want to make it more of a negative connotation of what's currently happening. I'd like to counterbalance on where we see glimpses of resilience that are basically being witnessed. Around the country, but equally on a global level. Now, as I probably have made you elicit from this, our new industrial policy is shifting from physical production to learning codes. Countries are no longer fundamentally competing on GDP alone. This is a game that has been played successfully for 80 years. On the onset of the end of Second World War, and it has brought enormous amount of prosperity around the world. No matter whether we're looking at the most aggressive growth country like China or the most modest one, Sub-Saharan Africa, every country on Earth regardless of where we are have witnessed some form of growth in the last 8 decades. What we see today is if the narrative is no longer about GDP, it's becoming more about learning systems. And so learning systems are coming with an incredible amount of opportunities because it does bypass what could be physical constraints of production. But it also creates a different kind of dependency. And so where is Australia in this conversation? Some of the work that I had done in preparing for this conversation, I found that Australia is currently entangled in a dilemma. You are capable of exporting incredible inputs but you're currently importing a lot of expensive outputs. And the problem with this model is that you are no longer capable of controlling directly the fluctuation of the dynamic evolution of the cost of import, especially in times where our supply chains are experiencing incredible amount of fragility because of the vulnerability that we have in, on the global system becoming reshuffled and reassessed almost on a daily basis. The question that I'd like you to start bringing forward in your universities is if we have to shift the paradigm no longer in the traditional industrial compound and the culture that we have been working on for so many decades, well, how do we start thinking of avoiding the mass media trap? And the trap will be, as you'll see later on, becoming a form of digital colony where you're largely dependent on the design of an architectural reference that is not necessarily defined here and it's not deterministic by the sovereignty of the country. It's defined by the large technology company that are becoming comparable to the oil and gas company but with significantly larger financial capabilities. When I try to look at where Australia was kind of ranking in terms of the investment in what we call the urban archipelagos, some of the highest concentration of innovation capital around the world, I found out that cities like Melbourne and Sydney, they are on the map. They're not necessarily where they could be. You notice the incredible difference between some of the cities that are currently dominating the new architectural reference. If you think about the innovation gap that still brings opportunity for us to reorganize and remobilize capitals, It's probably where the investment lag might provide a first level of analysis where maybe we like to consider how do we start bringing capitals to the country. They're going to close the gap between the input and the outputs, the way I mentioned before, so that you're building a learning system that is capital generating, the form of digital sovereignty that we've been mentioning before. Now, as I mentioned briefly, The trap is not able to necessarily navigate the tension across the opportunity in the short term versus the longer term view that are necessary for the infrastructure reference. A country might get entangled in a quicksand, which is kind of the middle space in which on one side we are largely dependent on technology companies are building their own national interests, as a matter of fact. On the other side, we might still have legacy systems that are creating traction on what have been working for us for a long period of time, but it will not necessarily have the same ability to capitalize on the value considering the fragility of the physical production. And so how do we build platform independence to some degree? How do we start bypassing supply chain fragility? How do we start avoiding the trap of digital colony? If we are not necessarily owning the systems, we are going to see a degree of dependency much bigger than just a supply chain dependency the way we have witnessed this before. But this is where part of my conversation wants to show you that the aurora is already visible. Australia is currently organizing itself across 3 interesting corridors. One is critical minerals. You have enormous amount of resources that are critical for understanding the integration between the physical and the digital infrastructure required. No software runs without hardware. The physical is not bypassed. We see this much more today as robotics is becoming again a critical function of production around the world. We do have a corridor on rethinking energy so that we are thinking of energy naturally renewable, bypassing the old system of reference that was largely driven by the supply chain ecosystem that today feels in shambles. And the third corridor is the one around using technology to augmentate or to advance physical assets. So the idea of biofood or agri-food that is coming with the integration of AI into physical capacities. And so Australia is a country that has the potential to move away from the impasse that many countries are currently to some degree suffering. Water is another critical part in the US, is at the moment poorly managed. In other country, it's managed better. Australia has so far been able to manage water as an opportunity to think of critical assets. In the work that we do, we see water become a form of wealth. The ability to manage water, it will become a competitive angle for countries, especially in times where international collaboration is not a given anymore. And we do see the rebalancing of coalition around the world. They're not necessarily following proximity, If you're looking at what happened in a country like Canada that had to strike balances elsewhere, it does show that sure, trade can be fluid, but anytime that we are trying to define some form of trade with proximity, we are encountering what could be the constraint of the confinement of the physical space, which is also largely determined by the geopolitics of the moment. And so within this context of having already glimpses of the Aurora in the country. Somehow the emphasis should go from national to regional. I'd like you to start thinking regional hub as living laboratories on how we can test and stress test multiple ideas. The climate volatility lab in understanding how climate resilience needs to be not only a way of mitigating or a zero-sum game, is rethinking the infrastructure reference considering the inevitability of the climate mutation. The other one is the energy grid that has to now become much more consolidated than what we have seen so far, where it was hybrid or complemented by existing legacy systems. And of course, food as a tech resilience, rethinking our food in the same way as back in the days somebody figured out that basically hunting animals was not the smartest way for us to survive. And we had to figure out agriculture as a way of bypassing the natural evolution of life. And so the focus I'd like to maybe bring to some of your attention when you're bringing this to the universities is maybe agritech and food systems. This is where we have seen a lot of really good experiment at the local level. The insight that I'd like to share with you is shifting much more to biomanufacturing. Today, the regional allocation of resources allows you to become more experimental. A testing hypothesis that you might want to see whether or not they create scale. And this is fundamentally the difference from the traditional economic concept of scale. We used to think about scale as a replication of some sort. Today scaling is kind of met with the scaling dilemmas where we do not easily replicate just because it worked. It's only the contextual adjustment and the ongoing reassessment on a local level that really help us to understand into what degree we can think about scaling. And so if you apply this to food or if you apply this to energy, you realize that this kind of innuendo across the clean energy and the cognitive energy is really where we can learn and eventually build hyperscalers. The language of strategy is fundamentally different. We start rethinking our foundations, not because we want to deny them, We simply want to expand them to the contextual zeitgeist we're currently living on. We are understanding much more the sovereignty is not just a luxury or a commodity that we want to have. It is something that inspires identity in many countries. It doesn't mean it does not come with reciprocity or global alliances. It just changes fundamentally that the assets are no longer coming from the assigned assumption that there is a benevolent sharing, but there is much more responsibility at the local level to navigate this degree of complexity across the corridor that I mentioned before between minerals, so production, what is energy and what is biotechnology, notwithstanding the fact that there's an overarching dome that is determined by the technology revolution we're currently witnessing. As I'm moving towards the last few slides of my conversation before some questions I'd love to entertain with you, I really invite you to start moving from state to system sovereignty. It's more about owning the nodes rather than protecting the borders. We are understanding more and more that this implies a cultural and a mental shift in many ways. And so to what degree we will be able to see this happen in the next few years in Australia, it will depend greatly on the ability to generate dialogues and insights in the university, inspiring the many people who are coming to this country to learn and to expand. So my takeaway message for all of you is building a cognitive capital is imperative. You do have the resources. The talent, and the value. We need to have the will, which is both functional and political, to build this cognitive infrastructure. When you build in sovereign cloud, which is the technology compound, when you're investing in regional hubs, and when you're closing the innovation gap, somehow you are redefining the future, not necessarily deterministic by the functioning of the existing trajectory, but it's really about building a future that does not exist necessarily yet, but that we can can imagine. I look forward to the conversation. This is my email if you ever want to be in touch with me. I thank you for having been so kind to welcome me to your beautiful country, and I look forward to a few minutes of questions. So over to you, Caroline. Thank you.

Well, what a great way to start today. Real challenge of different types of future that we could imagine and a clear role for universities within them. So we've only got about 10 minutes. I'll get started with some of the questions we have from the audience. First is, how do we require AI companies like ChatGPT to stop using freshwater to cool their towers and help the general public understand the damage this has caused?

Wonderful question. You know, in disclosure, I trained OpenAI model for 9 months before as a Caroline was introducing before, I serve as the chief economist in a Silicon Valley company, and I got into this because I started to train models. And that's when I realized that the environmental footprint of these models is hardly debated. Water, energy, and CO2 are currently subsidized by the desire of the usability of the technology. And I don't find that we're contrasting the efficiency of the model or the inefficiency, depends on, you know, how advanced you want to see that against the environmental footprint. I think it's a narrative about public discourse that needs to be much more upfront. But I like us to start thinking that we do not pay the price now because the end game is not to charge, is to improve the access. We currently estimated 1.2 billion people with access to some form of AI. The moment this is becoming 4, 5, 6 billion people, then I think the environmental footprint will become critical. I have seen companies that do not have the resources to design their own AI, but they had the resources to decide how to spend their money. And so there is a clear space where we can use digital dashboard to define the optimization of the models. I have seen this in country like Japan done with electricity grids. We can think about the same, um, when you're testing different model, you realize that, uh, the range and the spectrum between company like ChatGPT all the way down to DeepSeq really to redefine different level of, uh, energy efficiency. It becomes deliberate if we say this is not just about efficiency from an economic perspective, is also the efficiency of sustaining the model. And I guess this is where we have a role as educator to bring it into the equation and kind of change in the way we assess value.

Value. A question here, interesting, when talking about sovereign borders and so forth, how can Australia better leverage our close geographic proximity with Asia to power future growth? It's going to be relevant.

Yeah, it's a beautiful question. Look, I think it's inevitable that the future is Asian if we're looking at demographical tendency. Australia naturally leans in this direction geographically. I think it's the is a closer proxy to something that we consider regional integration. I guess though, the level of integration so far was mostly on outsourcing some of the manufacturing and benefiting greatly from a cheaper cost of production that was allowing us to expand, have access to goods and items at a fraction of the price. The narrative is changing when countries are catching up and they're becoming, labor costs more expensive. So the question is, in times of kind of building learning system based on what I mentioned before, owning the nodes, how do we work with these companies? There is enormous amount of talent, there's enormous amount of data that I think we can train. There's the ability to import talent that is going to distribute the domain training on the data so that we don't have necessarily a general AI, but we have an AI that is the determined by the needs of the given country. And I think using the proximity also to rethinking a regional integration that is bypassing the dominance on the Silicon Valley oligopoly.

Indeed. And, and perhaps leading on from that a little bit, not being platform dependent on firms like AWS and Microsoft is easier said than done in a country with the GDP and population of Australia. What would it look like for us to achieve sovereignty here?

That's another great question. I don't think we can break down this company and assuming that we can build our own. Those companies in Silicon Valley are an extension of a cultural norm that defined the Valley in the first place. The question is engaging with this company by having proactive governance requests that are allowing this company to work with. Look, in fairness, many of these companies, they have a benevolent posture. It's just that when we are unclear on the objective to achieve, the commercial interests, they speak louder. But I find that there's a lack of reciprocity across partners. In countries that I have seen managing this fairly, even smaller than Australia, like the UAE or even Qatar, it was really an engagement in working with this company to create or to co-create a charter of, I think, accountability. So it's really seeking those relationships. I think it's naïve to imagine that we can either break them down or that we can have our own technology companies because I think we are decoupling the fact that technology companies are an extension of a cultural normative process that happened in California. You saw this also in San Francisco being unmatched in terms of innovation. Capital, right? So I think this is the probably the most progressive way to, to look forward to that.

Well, perhaps talking about companies that don't necessarily have such a benevolent presence, social media has had unintended consequence with diminishing of discourse. And we had some really good sessions on that issue yesterday. Are we prepared to manage the unintended civic consequences of AI, and how should we prepare?

I don't think we are ready. I think there's still a lot of demystification that needs to be in place around technology that is not intelligent but is mimicking cognitive systems. And so I think they start from that. I do commend what Australia has done with the banning of social media for teenagers. Spain followed. And so you notice how learning systems, they can learn from each other quite quickly. I think that the question is rethinking social media by removing the neutrality of the access as a form of expression, but there's a lot of manipulation that I think is borderline between persuasiveness and manipulation. We see this in some of the research where technology is becoming a proxy for intimacy, and there is definitely a narrative about how do we improve this, but I think there's also a narrative how do we preempt it before it happens, and many of this time is because we don't see that the source our social fabric is being impacted. So I think technology accentuates, to some degree exacerbates the problem, but I find the root cause are more of some form of societal disparity that's becoming much more palpable.

And given that we're a conference about universities, we have a question here about how institutions, and I might say including and perhaps particularly universities, can become more agile to adapt to changing megatrends as constant change is now the new normal?

So I'll start by just sharing a personal anecdote. My training is in economics and I did my postdoc at Harvard Business School with Professor Michael Porter and stayed at HBS for about 10 years. I saw an iconic professor fundamentally disrupting the narrative about economics and strategy because he was connecting to the real world. So Porter was a mentor that taught me that what defines legitimacy is your ability to be relevant. And I think the question is, how do we be relevant in the academic world? It doesn't mean bypassing scholarly publication, but it means that a scholarly publication can be converted into a dialogue that is inspiring stakeholders, much more applied learning. I love to imagine that we can be much more lab oriented, experimental. Rethinking university as ecosystem of experimentation, I think, is really where many of the things already happen, is how do we make it institutionally even more prominent so that we are kind of even assessing contribution and value not only in formal systems but also in systems that are difficult to assess with the metrics of the past. I guess the invitation for all of you is, are you measuring value for what the 21st century asks you to measure? And if the answer is maybe, we're still very dependent on 20th century metrics. It's an opportunity to rethink in value in a much more multidimensional fashion.

Fantastic. Well, one last question, a couple of minutes. I might combine a couple of questions here, one of which just says, could you please talk to our government? But perhaps articulating that a little further, governments are conservative and often not planning for the future. How do we help them make the shift?

That's a great point. Look, I find government were never legitimized to think about the future, especially in democratic societies, because we are worried with 4 or 5 years, and there's a lot of tension across, you know, say, the parties in many ways. I guess it's shifting where political and technocracy might eventually strike a balance. I find it in country they're not democratically elected, so they don't necessarily spend too much time worrying about that. Technocracy is much more prominent, so hiring people that have expertise where the concept of the future is not a political ideology, is a functional trajectory in many ways. And so how do we integrate specific function in the governments that are coming from a point of expertise? And I'm gonna give you an example. I have a friend who happened to serve as Secretary of State in Spain Spain during COVID but is an academic like us, and he set up a center on foresight and geopolitical risk inside of the Prime Minister's office run by another colleague of ours, and that became one of the most successful functions that the government was running. So how do we start recognizing that the government should speak a language that is reflected on the extension of today? Definitely by also encompassing the fact that we can't establish success and KPIs in the same way that we have done in the past. So finding this kind of multilateral balance between political versus technocracy is probably a way to embed in the government narrative what government by themselves cannot easily achieve if they are overly preoccupied by political tensions.

Fortunately, our governments are never overly preoccupied by political tensions. So we won't have that problem. Mark, thank you for a really dynamic presentation, a very stimulating Q&A session. You've given us a huge amount to think about, and that is the point of this summit. So thank you for getting us off to such a great start. Thank you for having me. Thank you.