Fireside Chat “In Conversation: Sovereignty, Growth, and the Algorithmic Age: Greece and Europe at a Turning Point” at the Delphi Economic Forum XI
22/04/2026 - Articles & Interviews
With the Bank of Greece Governor Yannis Stournaras and Philippe Rogge, Worldwide Public Sector & Sovereignty Leader, Microsoft and Periklis Dimitrolopoulos, Editor-in-Chief, To Vima newspaper
P. DIMITROLOPOULOS: Good evening, ladies and gentlemen. We have not yet AI moderators, so I'm the human who poses the questions and runs this conversation.
So, the topic is about productivity and AI, and this connection between the two topics. We meet at a moment where the direction of our economies is being shaped not only by traditional micro-economic forces, but also increasingly by technology, by data, and the speed at which institutions, public and private, can adapt.
So, productivity, investment, resilience and human capital are no longer abstract policy themes. They are becoming the defining factors of competitiveness in what many already describe as a new algorithmic era.
So, we have to explore various items about this new reality, and I would like to welcome Mr. Yannis Stounaras, Governor of the Bank of Greece, who has provided one of the most authoritative macroeconomic perspectives on the country's trajectory and the structural drivers of growth.
And Philippe Rogge, President for Worldwide Public Sector at Microsoft, who brings a global view on how technology, and especially artificial intelligence, is reshaping productivity, public sector effectiveness, and economic competitiveness at scale.
Let me start with you, Governor. My first question is, you flagged productivity as the key to Greece's long-term convergence with the Eurozone. What are the main structural barriers still holding back, and where do you see the biggest opportunities?
Y. STOURNARAS: Well, thank you very much. It's a pleasure and an honor to be with you today.
Productivity growth is very important. It's productivity growth that, at the end of the day, makes an economy or a continent or the world resilient and also increases the welfare of people. So. it's welfare growth, it's productivity growth that brings all these benefits, something which we tend to forget. It's very important to have productivity growth.
What determines productivity growth? According to theory but also practice, it's two things, mostly. First of all, it's investment, productive investment. And second, it is reforms. That is, reforms that increase the potential growth of the GDP of every country.
Greece is doing well as far as investment is concerned, despite the legacy of the crisis, which destroyed capital, brought investment to a very low level. In 2019, so many years after the crisis, investment was 11% of GDP. Now it's 18% of GDP, but it's still four percentage points of GDP below the European average, which is 22% of GDP.
So, we catch up with investment, but still there is an investment gap which we should also try to achieve, to fill.
Second, reforms. Greece, despite the fact that labor cost competitiveness or price competitiveness have improved tremendously, because of the measures that we took during the crisis, still in terms of structural competitiveness —and I will explain what I mean by structural competitiveness— ranks rather low in the world.
Why? Because, let's say, of -still- problems in infrastructure. Second, because of delays in justice, despite the progress recently. Justice delays in Greece generally, it is not in the international ranking we are below the average in terms of speed of justice, of decision-making in justice.
Number three, I would put the mismatch between labor demand and labor supply, which has to do with the education system, which does not produce the skills that the labor market wants. I would say the demographic problem also is very important, but I will stop here.
Of course, now, as you say, we are in the era of artificial intelligence. This is extremely important. Education plays a huge role. In my view, it is education that can solve these problems. Digital education, how to educate people to learn themselves later when they have aged.
So, to me, it's extremely important that AI is not a frontier technology, but it is dispersed in the society and in companies.
P. DIMITROLOPOULOS: Yes, but, Mr. Rogge, in Europe, there is heavy investment in digital. Okay, but productivity isn't following. So, what's wrong?
PH. ROGGE: First of all, thank you for having me here. It's obviously always a pleasure to come to Greece and to come to this forum and to be on the panel with both of you.
It may sound like we've rehearsed this, but I'm going to complete your sentence, Governor. Real benefits and real gains start not where technologies are invented, but where they are being used. If you look at the printing press several centuries ago, the Germans claimed to have invented it. It was their neighbors who ran away with it and started basically printing books first. Obviously, they were overtaken, but the Dutch and my home country, the Belgians, would claim a moral victory over the Germans who actually invented the printing press.
When you look at what's happening in AI today—and I'll focus on Europe—you look at Norway, Ireland, France, these are countries who today are doing a number of things. First of all, they're investing in their own LLM models. Secondly, they're trying to apply AI, not just LLMs, but AI to fundamental research. Thirdly, they're investing heavily in skilling, which you mentioned, which is really the foundation that human capital is really the foundation of any progress. And then finally—and this is where I have a bit of a caveat—they're investing in infrastructure.
I believe—and we believe at Microsoft—that we need to be thoughtful with investment in data centers and in infrastructure. They need to be part of the mix. They are not the panacea to all problems.
And I say this because if you look at the two countries who today have over 80% of the world's data centers, the US and China, their digital usage index is between 20 and 25%. In the Middle East right now, the UAE, or you take Singapore, the overall index of usage of AI is in the ’60s.
So, the diffusion of the technology is to be seen as disconnected from the actual investments in infrastructure, which are important but should not be misled or misinterpreted.
The good news for Greece is that if you look at the youth, Greek youth between 19 and 24 years old ranks number one in the use of generative AI in the EU. I think that's a huge opportunity. That's something to hang on to.
Clearly, that usage drops off in the in the next tiers of age, but clearly, with the youth being so locked into the use of artificial intelligence in Greece, I believe there's a real opportunity. And that is what ultimately is going to be the driver, the engine, if you will, for productivity.
I'm sure you have calculated in your latest report what the impact could be for Greece. When you look at the IMF, the IMF calculates at least 1% of the next five years’ productivity gain for the EU.
Y. STOURNARAS: Annually?
PH. ROGGE: Annually, yes.
Y. STOURNARAS: 0.8% annually.
PH. ROGGE: I'm rounding up because we're trying to create optimism. If you look at the more pointed use cases where, you know, we go in, we have our own, let's say, research institute, and we go in, and we basically calculate reasonably scientifically how more productive white-collar workers—of course, we're fine-tuning—you can get up to 40 to 60 hours a month gained.
So you're starting to get significant advances. And we have customers who have come back to us and basically said, "For the right use case in the right customer, I can get up to 30% for these specific workers using AI."
So, I think the potential is there, the studies prove it. The question is, how do we diffuse it, and how do we put it in the hands of everyone.
P. DIMITROLOPOULOS: So here in Greece, we see that business sentiment is improving. A series of surveys from the Bank of Greece show that. But, still, they struggle to compete intentionally. So, what do we need to do to improve this situation?
Y. STOURNARAS: AI is extremely important. We have many small and medium-sized enterprises in Greece. It is not important only for the large companies to adopt AI. We should try to diffuse AI to smaller companies. They become more digital. The public sector also becomes more digital.
Greece has done miracles, starting from the crisis when more than 80% of analysts and academics and some politicians thought that Greece would not make it in the Euro. Not only did it make it in the Euro, but now it's growing twice as much as the rest of the Euro. The public sector debt as a percent of GDP is falling very, very rapidly. So, it's a miracle. But the question is, how do we continue from now on? Technology is important, but we should not forget that the number one is fiscal stability, second is financial stability, third is to continue the reforms that we are doing now.
Artificial intelligence: There are many scientific articles written. There's no doubt that productivity will go up. As you said, the IMF and the OECD, they say it's 0.8 to 1% annually. Huge productivity growth.
Don't forget the Solo Paradox. Solo was an excellent Nobel Prize winner, perhaps one of the best qualified economists ever, who said that productivity is everywhere except in statistics. Why? Because it is not important to have a discovery, but we should train people to take into account this discovery in their everyday work.
So, the education of people is extremely important if you want to compete with others, I think. So that's why I'm saying education, education, education.
Now, to go to the more, let's say, trivial things like the current account deficit, etc., as I said, despite that Greece has increased its export ratio from 20%, now we export 40%, not only, as people think, that it's only tourism and shipping. It's not only that. It's also now industry, it's pharmaceuticals, it’ s cement, it’s cables everywhere.
So, Greece has changed completely. But we also should try to substitute some imports. Import substitution is also important. But I want to say that exports is number one, I would say. We are still a long way to go to compete with others. By investing mostly in education, I would say.
It's not a question only of efficiency, but also of social justice because equal opportunities are important. I mean, to have an elite society is not important. We have very good private schools. We have very good scientists, all over the world. Greeks are the best scientists all over the world. But the question is not that. The question is how to diffuse, how to create justice giving equal opportunities to everybody. And this requires good state education.
P. DIMITROLOPOULOS: So, Mr. Rogge, as a Microsoft man, you have worked with governments globally. And the question is, what's the most common mistake they make when using technology to drive growth, the governments, the states? Is there a mistake they're doing?
PH. ROGGE: Well, I prefer to talk about the good examples, versus the mistakes. And then you can deduce, you know, the ones who maybe miss a step.
P. DIMITROLOPOULOS: Or about the practice.
PH. ROGGE: I would say the good practice is our governments or public sector organizations who do a couple of things all at the same time. First of all, they don't just use the technology. They use the opportunity to shape and to leverage, in this case, AI for, let's say, advancement of the country. Whether it's, you know, the domestic advancement or even in the different rankings, you see specific governments, specific countries pick up AI right now as an opportunity to leapfrog other countries and just increase their competitiveness. Wealth of nations etc. That's number one. So, they are shapers. I've read with interest, what the special secretariat for foresight in Greece has written. I think it's well done. Of course, now it needs to be executed, but I think there's a level of perspective there, which I find encouraging and positive.
Secondly, they use. They realize that they are customer zero. The public sector has in many countries a significant overhang in the economy. So, that's the first factor. The second one is, by using it, they can improve, of course, their interactions with the citizens, but they can also stimulate small business, large business, export, trade barriers, you name it, by, let's say, improving the way they license, they pay, they procure, the bureaucratic process, the speed and quality of decision-making. All these things impact Greek businesses wanting to do business domestically, potentially tourism or international trade.
I think the second one is their active users. And the third one is they regulate. And when I say “regulate”, for me, it's more than, writing some laws. Of course, that needs to happen, but I think creating an environment of trust is key. And I want to piggyback on what you said, which is that trust of course, you know, there's elements related to sovereignty; there's elements related to cybersecurity; there's elements related to the fidelity of results of these AI technologies. But there's also elements related to social inequality. And the only way people are going to massively adopt these technologies is that they feel that they can trust what's happening. And if you create, to your point, a too big disconnect between the haves and the have-nots, then it turns into a downward spiral.
So, we need to build that trust. And you see governments around the world actually embrace that and try to make sure that because of this diffusion there's a level of equity in how it gets dispersed.
P. DIMITROLOPOULOS: We have only two minutes, so, I want a last comment.
Y. STOURNARAS: I will stay with AI because it's important. As I said, definitely we're going to have a productivity increase, but, cui bono, who's going to benefit from that? If we create AI that substitutes people, then only capital will gain, labor will lose and we're going to have social problems because people cannot stay at home and live in subsidies. If AI complements labor, it's going to have a new page for humanity, I think.
And if we take into account also the development of quantum computers, which is another invention, the combination of the two could also reduce the energy used by AI. But that requires public institutions, government intervention, so that there's diffusion and complementarity rather than substitutability.
PH. ROGGE: I would probably add, I’ll take a risk here, and I'll add a perspective on Greece, if I may, and I hope to be able to leave the room in peace and quiet. I think in terms of overall diffusion of artificial intelligence in Greece, there is work to be done. As I said, I'm very hopeful about the penetration of these technologies with your youth and so please embrace that.
I think the next step is skilling. As a company, we're deeply invested in continuing to skill and supporting any program, that can help because ultimately that also leads to that equalization and that equal access.
And once that's there, then I think it's really going to be about the public sector. If the public sector leads the way, if the public sector basically is the pioneer in this space, then the whole economy, the whole country, not just the citizens, the small, medium businesses, and every international company will basically benefit because that rising tide will lift all the boats.
I am actually quite optimistic when you look at the DESI index, which is sort of the index of European countries around their digital maturity. Greece has a lot of upsides, let me put it that way. But the one area in which they really outshine many of the other countries is in their public sector adoption of technology. So, I want to applaud the Greek public sector, at the same time continue to encourage them to continue to invest because I think this is the key to unlocking the full potential of these technologies for Greece and in Greece.
P. DIMITROLOPOULOS: Thank you very much, both of you, Mr. Stournaras, Mr. Rogge.
Y. STOURNARAS: Thank you.
PH. ROGGE: Thank you.