Open Source Program Offices for Good - UN Open Source Week 2026 (Part 6)

Alright everyone, okay, we're gonna get started and I want to introduce Christina Goodness, who will be our moderator for today for this panel. She's a special assistant to the UN Chief Information Technology Officer and head of the front office. So please welcome Christina Goodness.

Okay, thank you. Thank you for your patience. Just checking, how are we doing? I know this is the end of the day, but I think we have a really excellent panel that I really look forward to walking you through. As a first step, we're very, very honored to include Roland Molleros. I hope I am saying this correctly. The Director of the UN Office of the High Representative for the Least Developed Countries, Landlocked Developing Countries, and Small Island Developing States. That's an excellent acronym for you. Please, please join me in welcoming the speaker to speak on the work of that office and open source. Please.

Many thanks and good afternoon to you all. Excellent introduction, and UNOHRLLS is that acronym of the Office of the High Representative for Least Developed Countries, Landlocked Developing Countries, and Small Islands Developing States. UNOHRLLS. Excellencies, distinguished delegates, colleagues, I'm very pleased to join you for this very important discussion on open source at the United Nations. I very much thank the organizers for convening this timely dialogue and for bringing together the UN system, member states, and the global innovation community around a shared vision of practical digital cooperation. We meet at a moment of profound technical change— technological change. Science, technology, and innovation driven by rapid advances in digitalization are reshaping how we deliver our mandates. From artificial intelligence to data-driven policymaking, these transformations are expanding our ability to serve more effectively and reach those furthest behind. At the same time, they are requiring us to rethink how we collaborate and uphold our core values of inclusion, transparency, and accountability. This transformation is unfolding unevenly. For the least developed countries, the LDCs, the landlocked developing countries, the LLDCs, and the small island developing states, the SIDS, that UN-OHRLS serves, the digital divide remains a persistent and structural challenge. While digitalization holds immense potential, access to its benefits remains limited. The data illustrates this gap clearly. Only 34% of individuals in the LDCs and 38% in LLDCs are using the internet. SIDS perform better at 64%, but still remain below the global average of 74%. Internet access at modern 5G speeds remains constrained. Only 5% of people in LDCs, 9% in LLDCs, and 25% in the SIDS have access to 5G networks, compared to a global average of 55%. Affordability also continues to constrain meaningful connectivity. These realities underscore a central question: how to ensure digital adoption in ways that are inclusive, affordable, and nationally owned. Excellencies, it is in this context that open source emerges as a compelling enabler. Open source provides a model of cooperation that is at the heart of what we do at the United Nations. It promotes transparency, reduces costs, and allows countries to adapt and scale solutions based on their own needs. It also strengthens digital sovereignty by giving governments greater national ownership over their digital systems. It supports interoperability and collaboration, allowing institutions to deliver global public value more efficiently. This vision is increasingly reflected in system-wide efforts, including the UN Open Source United framework, which marks an important step forward embedding openness, collaboration, and reuse at the core of our digital transformation. We at UNOHRLLS, we are working to translate these principles into practice. The SIDS Center of Excellence, launched in 2024, promotes knowledge sharing, partnerships, and technology cooperation, including through its technology and innovation mechanism. In parallel, work is advancing towards an online university for LDCs, as called for under the Doha Programme of Action. This initiative aims to expand access to higher education at scale and to address the tertiary enrollment gap in LDCs, which stands at just 11% compared to a global average of 40%, with only 21% of graduates in STEM disciplines. We are also collaborating with partners, including UNECE, to support LDCs— LLDCCs and SIDS in leveraging UN open source digital public goods to facilitate trade and reduce trade and transport costs. These efforts demonstrate how shared, open, and collaborative digital approaches can help overcome structural constraints by pooling resources, lowering costs, and expanding access to knowledge and innovation. Excellencies, looking ahead, the priority is to move from principles to implementation. Allow me to highlight 4 key directions. First, strengthen regional cooperation, particularly for SIDS, LDCs, and LLDCs, to pool resources, share expertise, and achieve economies of scale. Secondly, promote open and interoperable digital public goods. Public sector digitalization must be sustainable adaptable, and inclusive. Third, invest in digital skills and capacity building at scale. Partnerships across governments, academia, and the private sector will be essential to equip people with the skills needed to participate in and shape the digital economy. And last, fourth, address affordability and access barriers. Innovative financing, targeted policy support, and international support will be critical to ensure that no country is left behind. These priorities are fully aligned with the Doha Programme of Action for LDCs, the Awaza Programme of Action for LLDCs, and the Antigua and Barbuda Agenda for SIDS, all of which recognize the transformative role of science technology and innovation in advancing sustainable development. Excellencies, in closing, open source is no longer a niche or peripheral issue. It is becoming central to how the United Nations can deliver more effectively in a connected world. For the countries we serve in our office, it offers a practical pathway to bridge digital divides strengthen resilience, and accelerate sustainable development. The task ahead is clear: to translate this potential into tangible, scalable outcomes that respond to national priorities. I thank you very much.

Okay, thank you, Roland. Good afternoon, everyone, and welcome. On behalf of the Office of Information and Communications Technology, one of the organizers, co-organizers of this Open Source Week, I am delighted to welcome you to this panel on open source at the UN. We really just could not pass up the opportunity to share the work that's going on inside the UN. As Bernardo mentioned earlier this morning, we must walk the walk, not just talk the talk. Todaí, you'll hear about the innovative work going on in the UN Secretariat, in the agencies and the funds and the programs. You'll also hear about some of the challenges that we face. These are not small challenges, as our colleague just said, but hard problems perhaps are some of the best to solve and most worth solving. Importantly, I hope that we can share how we in the UN and members of the global open source community can continue to work together. It's not a new conversation, as many of the speakers have said before, how we continue to work on global challenges. The UN Secretariat has long trusted and relied on open source technologies to power our most critical infrastructure. This is not abstract. For many of us on an individual level, our teams and our organizational successes have been built on the foundations of open source. Including my own, my very own use of Berkeley Unix at age 17. Yes! The work of the UN has long leveraged open source components, and this is thanks to you, the open source community. At the same time, we are in a very specific moment in UN history. We're working very hard to strengthen collaboration across the system by co-chairing the UN Open Source Community of Practice, which brings together colleagues from across the UN family to exchange experiences, coordinate efforts, and as Bernardo said, to go further faster and advance the adoption of open source urgently across our organizations. I'm very much looking forward to moderating today's discussion, and with that, let me turn to our distinguished panelists. Give me a moment. I invite the panelists to come to the panel. To the, to the stage. Uh, first I'll introduce Mr. Guillaume Canella, the Director of the Division for Digital Inclusion and Policies and Digital Transformation at UNESCO. Then we have Shireen Hamid, the Chief Information Officer of the International Monetary Fund. Vincent-Jean Yves Martin, the Director of the Office of Innovation, Food, Agriculture Organization of the UN, FAO. And last but absolutely not least, Mumtaz Tamim, Director of Digital Impact, Technology and Innovation at the UN Relief and Works Agency for Palestine Refugees in the Near East. Please join me in thanking and welcoming the panel. Okay, testing. Okay, thank you. We will start the discussion now. Let— maybe let's start with one question. Here we are in 2026. AI has supercharged the open source environment, the technology environment at large, and we have 4 entities here, 4 very different mandates. Maybe let's start a question for each one of you in turn. If you go first, Sharon, what does open source actually make possible for you? Um, are there examples you can share quickly on open source and transformation? Please.

Well, thank you very much and delighted to be here with all of you. As you know, the IMF's mandate is to promote global macroeconomic and financial stability and support to its members through policy advice, financing, and capacity development. So from the IMF, open source makes it possible to move from sharing information to enabling capability. Delivering value and impact. And we've seen this evolved. We've evolved from open data, which is key to making trusted macroeconomic data accessible, to sharing analytical tools like Growth at Risk, and now to AI-enabled tools such as ChatGPT that allows countries to interact with data more directly and adapt solutions to their own context. Ultimately, it enables transparency, adaptability, and productivity, giving the ability for institutions to build on each other's work rather than starting from scratch.

Thank you.

Good afternoon, everyone. Pleasure to be here. So UNESCO is a very complex organization, as you probably know. It's a different areas of mandate. So we are the UN organization for education, for sciences, for culture, and for communication and information. So we— the open source discussion is both an internal interesting challenge, but also we have the mandate to advise member states on those topics. I'm not on the internal side of the conversation. I'm this side of advising on this topic. Very briefly, a few things that are part of this conversation. In the very first paragraph of our Constitution, we have that UNESCO has the mandate to promote the free flow of information and ideas. And as you can imagine, to achieve that goal, one thing that is essential is to guarantee the openness of the systems. Otherwise, you can't do that. So a few examples in our areas of mandate. For example, we have a recommendation on open science. And you could see during the pandemic how much open science was essential to reach the vaccines in no time. So if we want to demonstrate the actual impact of openness, this example has a massive demonstration of impact. On education, we have a specific legal document called the Open Educational Resources that some of you referred during those two days. And you can see examples cross the border of how much governments have been saving only to not need to pay more proprietary textbooks in adopting open educational resources in their procurement services. Another area that we are in charge is the protection and the stimulation of journalism. So if you see cases of investigative journalists all over the world impacting on their investigations, like for example on corruption or human rights violations. By using open data sources and datasets, again, you can measure the impact of the openness in the system. And I could continue. Also in the free and open source element, you can see how governments have not only saved in terms of procurement in using open in free software rather than proprietary, but also in terms of digital sovereignty. So in a nutshell, we are trying to show to our member states and advise them on this value of, at the end of the day, protecting the free flow of information. Thank you.

Thank you.

Yes, so I'm from the Food and Agriculture Organization of the United Nations. I must say that there have been a real paradigm shift in FAO for the last 4 years when we've been moving away from developing individual digital solutions toward what we call this DPG-first approach. And I must say that these days we are also very much focusing on the broader digital ecosystems approach to make these solutions more adaptable and scalable. But to answer your question very precisely, when you say, what does it make actually possible? So from my perspective, what it makes possible, it makes democratization of science and innovation possible. What I mean by that is that thanks to open source approaches, we've been able to close the science, technology, and innovation gap and bring the solutions, in my case, to smaller farmers that most need these kind of innovations. I can give you two examples. The first one, which is more at the global level, we've developed this Open Forest platform, which helps us track— which helps member states track forest management, environmental changes, and also to report on their climate commitments. This platform, which is open source, which is one of our 11 DPG in FAO, is being used over the world by 250,000 users now, and it's helping countries to make their report on their climate commitments. So that's more the global national level. But now, at the local level, when I say closing the STI gap, we have what we call the Digital Service Portfolio. So this Digital Service Portfolio, which is also a DPG, which is open source, is really bringing technical knowledge to the farmers in terms of climate, in terms of markets, in terms of livestock production. And thanks to having access to this information, they can take informed decisions and improve their livelihood and production. So we've got this— I mean, this shift in mindset in FAO, using more and more DPGs is becoming totally natural and promoting open-source solutions. And thanks to that, we've been able to close the science, technology, and innovation gap.

Gap.

Thank you.

And Mumtaz?

Yeah, thank you. From UNRWA perspective, and UNRWA is an agency specialized for providing relief and work agency, mainly for the Palestinian refugees, so we are like a small government providing services to a specific population covering education, health, and relief. And often other sister organizations call us like a mini-UN because we somehow do what UNESCO does and what World Health Organization does and what World Food Program does, et cetera, but for a specific mandate. So in our journey toward open source adaptation, we can give a few examples. I would refer to one which is much more recent one on providing a tool for running the test for Palestinian refugees in Gaza where they've lost the schools, but yet they need to pass an exam. So without going into software development cycle or without going into procurement process, we picked up a tool just from off the shelf, from open source,. And in a matter of one month of work, we prepared this platform. We teach the teachers how to put the test, and we have run the authentication process for the students to do the exams. And it has to be so simple in a way that even the kids using the phones of their parents, they can run the exam using the authentication of their student ID and their date of birth. And we managed to run this test successfully for around 230,000 students. And this was done in less than a month of work. Actually, this improvement shows that the impact was so quick. And the data and the correction of the exam results were done in a matter of hours, not days. And the data were linked to our enterprise the education management system, and the student got their marks as well done all automatic. So this was a major achievement for how fast you can use the open source technology in, in this regard. The second example I would give, and that's a very exciting example, is on the software development cycle. You will see that there are plenty of tools now exist in the market AI. It's an open source AI on application development, and it's not only for writing coding, but it's for the whole lifecycle from business requirements to mapping to coding to testing to user acceptance testing. All that lifecycle is in, is in that ecosystem. So those are the two examples that I could share with you from, from Onward perspective.

OK, there you have it, some very concrete examples from forest management to education and testing, and really serving the very challenging mandates that the UN has in these different agencies with very— let's just say resources are— we are eager to find resources, locate resources. Maybe I'll turn back to Guillaume. In UNESCO, you mentioned the work on open science And as it relates to open source and the promotion of the work in UNESCO to promote the word open across the board and understand how the world's knowledge really should be continued to be championed to be open so that we can— further generations can benefit from it. Maybe take us back. How would you say UNESCO's approach to open source has evolved over time? Is there a commentary, especially considering, you know, collaboration either within UNESCO or across other organizations? How has that evolution gone?

Thank you. First, it's, as we all maybe realize, it's not easy to be open and transparent, right? I'll tell you a story. This year, coincidentally, we are marking the 260th 60th anniversary of the first freedom of information law in the world, that is the Swedish law. This law was approved in 1766. The second law on freedom of information was approved only 200 years later, in 1966. So you can see how governments— and we are an intergovernmental organization— are reluctant in being open and transparent. And to continue the anecdote, in 1990 there were only 10 countries with this kind of legislation. Good news, today there are 145. So in the last 30 years we had a super expansion in this, at least this political willingness for the openness and access to information and open of knowledge and so on. And then we have the World Summit of Information Society and this and that. UNESCO also evolved through these same waves of changes. So, for example, UNESCO was in charge of the Universal Copyright Convention in 1952. I don't know how many of you know that the seal of the copyright was invented in UNESCO. So moving from that background to the openness of open educational resources, Creative Commons, et cetera, is a big shift. For the organization and for its member states, etc. So I would say, Kristalina, that one of the key elements is precisely the ability and the lessons learned to demonstrate how much the openness of knowledge, of science, of education, etc., was central to the development agenda of our member states and in particularly to good governance. A chief justice of this country used to say that the sunlight is the best disinfectant. The more you have transparency— and this morning someone said that you have more security in many ways, and open source contributes a lot for that security for everyone— and you have more trust. It's a lot of what we are actually desperately in need right now. Right now to, to keep protecting and promoting in the system. Thanks.

Perfect. Thank you. Um, as you can see, the work takes many forms to promote open knowledge, uh, and open source is key to that. Maybe I'll turn to Mumtaz. The open source building blocks underpin the whole tech ecosystem, as many have mentioned throughout the day, including tools during crisis. UNRWA delivers to millions of refugees under very intense resource pressures. Could you maybe give us a couple of concrete examples of open source? You've mentioned one around the testing and education, but making something happen where it wouldn't happen at all otherwise.

Yeah, thank you. You know, the open source journey at UNRWA is evolving, so we picked We adopted the open source, you know, after, let's say, the web development, you know, platform that we had. But currently, we are moving away from— we are moving from using open source as a tactical decision or opportunistic decision to a more strategic dimension. And the reason for that is a lot, but I can give, you know, some examples. There where the geopolitical risk is real and you can see that a lot in the Middle East area where the tension goes on all direction. This is why the need for having solutions ready on the ground for business continuity and resilience is a must for us. The third point is about the vendor lockdown and the need for having that kind of independency and sustainability is a key. And hence, having all those tests of open source solutions for our ecosystem is much, much more needed. This will provide us the flexibility that we need and the independence that the UN agencies need to have in order to serve the whole humanity, because, you know, we should not be depending on certain technologies, certain vendors, certain country to deliver our mandate.

Okay, this is a little bit of a turn of the conversation that goes beyond some specific individual examples and really talking about the digital infrastructure at large, the productivity suite, the entirety of the work environment. It seems like there's some shifting going on. I think that's been alluded in a couple of cases. Maybe I'll turn to you, Sharon. And we're in this moment of change, right? As many have mentioned, turning to the adoption of new technology, as we've seen with AI, initial adoption sometimes is very easy. You know, you turn something on, but sustaining it is really hard. And how has IMF approach to openness evolved from data and tools to more advanced digital capabilities? And what did you do to kind of make it stick?

Very good question. And I'll tell you that, you know, in the context of building technologies that are durable, I don't think that's a single factor out there. You know, I think the core ingredient I look at it in terms of the top 3 elements that we need to move together. One is about strong governance and policy foundation. Here, we're talking about licensing arrangements. Mamta's talked about lockdown. We don't want to be beholden to one organization. We need to make sure that the standards and the approaches that we take gives us a variety of options for us to be able to deliver on our mandate and our objectives. At the same time, you know, we have to also look at data sensitivity, right? With data privacy and data being out there, we need to make sure that we have a strong policy in place to actually understand what are the data risk, and how do we protect the information and data that is being provided to the Fund for the use of our members. And then the third is to actually look at it from a risk management perspective. We need to make sure that in all contexts of how we deliver our, our technology and our platform, we look at it with an enterprise risk lens. And here, at the end of the day, you have to have an appreciation— the mandate of the Fund is about making sure they are on macroeconomic policy and at the same time financial stability. If we don't build technology that is durable, right, it has significant impact to country and the country's macroeconomic and financial posture. So for us, we need frameworks to make sure that we ensure trust, clarity, and at the same time consistency across how we use our capabilities. Second, I'll say it's about our institutional collaboration. To build durable technology, a lot of us think it's just about the technologies. The technologies need to come in, put the technology, and voilà, we have a good system for everyone. That's not purely true. To actually deliver a robust and durable technology, what we need to do is actually collaborate. And here, the collaboration is actually the village. And the village is what we're talking about, is ensuring your legal posture structure is in place, your business and the subject matter expertise are in place, you have the technical expertise in place, as well as what I call an alignment with the organization. It has to align with the organization's mandate. It has to align with the priorities of the organization. So we can't look at this in a fragmented approach. And then finally, in terms of sustainability, open source is not about building a technology, putting it out there, and hopefully somebody is going to use it, adapt it.

Right?

It doesn't work that way. Arguably, what you need to do is still continue to maintain, support, and evolve the technology over time. I will tell you that it's not an easy journey to make it durable, right? If you have a technology that is open source and you put it out there, but the solution actually depends on specialized infrastructure, talent, specific talent, or external systems, then technically, what you think is open is not necessarily open. Because it's not truly usable, right, or sustainable in many contexts. So more broadly, in terms of challenges, we still have what I call capacity constraints. We still have fragmentation that actually happens that, you know, remains an important barrier to us moving forward.

So there you go, community. These are some really important words, I think, because as we mentioned before, we're very— We're reliant and dependent on our collaboration with the open source community in the UN, and this is what it takes to make it stick. So hopefully this can be taken to heart, food for thought, certainly for all of us. Vincent, if I could turn to you, perhaps let's bring you in to speak to the field. We're here at headquarters in New York in this big, beautiful room, but what, from your background and from your experience, What was the single greatest challenge to implementing open source projects from your experience? I'm thinking about challenges maybe across regions, across agencies, and introducing innovations.

Yeah, yeah, and I would like to build a little bit on what Shirin was saying because I think it was very important to talk about these ecosystems around these solutions. But just before, I gave you two examples which were quite successful. I mean, one which was more at the at the global level, but how it has been used at the national level, Open Forests. And another one that has been deployed in several countries, which is our digital service portfolio. It started in 2 countries, now it's in, I don't know, 10, 15 countries, and it's accessible to farmers, and it really, I mean, in their own languages and so on and so forth, considering also the digital literacy and so on and so forth. But I could give you plenty of examples of where it failed, and we failed really flat. Even if it was open source, even if it was a DPG, it didn't work. And why it didn't work? And this is something I've seen everywhere I've been working, from Asia to Africa to Latin America, is problems of adoption and scale. And it was mentioned already by, uh, by Sherine and Guilherme, because we, we think because it's open, uh, it's going to benefit people equitably. 'But it's not. Because it's open, it's going to be used and adopted and scaled.' No. So the thing is that it needs to be accompanied by investment in skills, in governance, in capabilities, in capacities, and so on and so forth. So I think the surrounding environment, the ecosystems around these solutions, is more important than the technology itself, basically. And that's why now in FAO we are not so we are not so rigid on the being open and not being open. We believe in DPGs and so on and so forth, but we are not taking it in a rigid way, more in a flexible way. We first look at what is the problem we are trying to solve and what kind of solutions we can put into place by strengthening the ecosystems. So I've got 3 kind of examples I would like to share. So from the local level, how do we overcome this problem of adoption? It's not just— I mean, we had several techniques, but one of them is to use social sciences, including behavioral science, to really understand what are the barriers to adoption before starting anything. And then you do your intervention, and then you can measure the impact afterwards. But from a social perspective and social scientist perspective, you can look at it, and that's the way we prevent a lot of failures from adoption to scale. Scale. So the second one I wanted to mention also is, because here we are all UN agencies together, we don't talk much about what we do together, while I think it's really important that the direction where we should aim at. And last week we just launched a global initiative on AI for agrifood systems together with ITU, WFP, IFAD, and FAO. So I wanted to mention it because in FAO we already have our roadmap on AI for agri-food systems, but we believe that joining forces with other UN agencies looking at this governance mechanism, data sharing agreements, use cases that are important, is the way forward. So that's why we launched this global initiative. And finally, last point is We are going to— we are launching this year a project called Mainstream in Sub-Saharan Africa, and we are going to establish scaling innovation hubs in different parts of Africa with the idea of the main value that we bring to this project is strengthening these ecosystems of innovation.

Well, certainly some large challenges, as we were— as I was mentioning at the beginning. You know, the agri-food system open source across regions, across entities, that is certainly not a small challenge. But yes, the hard ones are perhaps the ones most worthy of solving. I also heard sort of strains of conversation around the responsibility of protection really extending to open source technology and technology stacks at large, and how we permeate these values into things that we build is really important, especially as we are interdependent, essentially, to tackling these global, perhaps, programmatic challenges, but then using the tools that we have at hand. I think we're coming up on the end of our panel, so maybe what I'll do is a lightning round, if you'll bear with me, a last word for each to each of our panelists, and maybe we'll start with you, Mumtaz, and work our way this way. In one sentence, you know, we build a lot of our projects in isolation. Um, hopefully that's changing. I think actually all of you did say it was changing, but if you were to turn to the entity sitting next to you or turn to any of the participating organizations that are in the audience, maybe what would be the one collaboration that you would ask the open source for the community both in and outside of the UN to work on from your perspective? And one sentence, one sentence.

Okay. Thanks.

If you can.

Thanks. You know, a few years ago, we had the priority number one was, you know, 100 cloud-first, right? So that was, you know, now we need to shift that into open source-first. First, where possible. And where possible because we need to be realistic on what we can achieve and the change cycle. It's not something that you can do overnight. It's an evolvement. And certainly, that's something we cannot do it alone, agency by agency. We need to have more collaboration across this and more research and development towards certain projects and activities to move this forward.

Perfect.

Vincent?

Yeah, I mean, I think this is what I was saying just before. Our challenge today is not simply promoting open source. We need to do the hard work now across organizations, across institutions, partnerships, and so on and so forth. So I think that's really the key message. And because you were asking what would you ask the community, I would ask them to join us, to join this global initiative on AI for agrifood systems.

A very worthy cause.

Coming—

growing up on a farm myself, a very worthy cause. And Guillermo?

Oh, I think that what Vincent said before about the meaningful connection of the real people is our challenge. Because even, of course, if open source is different of proprietary technology, at the end of the day, we were quite successful in connecting people to the grid, but still essential elements for guaranteeing this process is meaningful for them, like for example multilingualism, like the content is a content that makes sense to their own reality. In those areas, we are still far away of what is needed. So I guess it's a collective invitation because in UNESCO we also have these The issue is how we actually reach that last mile, but in a meaningful way. I think this is our collective challenge, including in the New Development Agenda post-2030. This is it. It's easy to say, not easy to do. That's why we need this collective thinking here to advance.

Very good. Lastly, Sharon.

I'm glad you say it's a sentence, because the sentence can actually be a long one. Here, what I would say I would say in terms of collaboration to— within the open source community is I'm going to look at it a little bit differently. I think it's moving beyond just sharing your code and your model. And here, move to what I call building the capabilities as a community. And when you build this as a community, make sure that it's trusted, it's reusable., right? And other organizations, if not countries, can use that with confidence. And in doing that, you have to do that by aligning on what I call a shared platform or platforms, right? Some common licensing arrangements that can benefit everybody. And then begin with always the end in mind. That's what my husband always say, right? When he proposed to me, he says, always begin with the end in mind. And this is by designing for reuse from the onset. Don't think that you're building an open source just for yourself or a specific purpose or an objective. But at the end, what do you want to get out of this? It's naturally easier to share and actually look on the onset what the barriers are. When you look about doing that social testing and understanding what are the barriers to entry, Go with that end in mind, and then so that you can actually understand how to actually fix the problem upfront rather than the tail end. Thank you.

Excellent. Thank you so much. I think, you know, looking towards the next 7 generations, as some people say, is really the goal here. And we really look forward to working together with you, the open source community. So please let us thank our panelists and— yes, thank you. Thank you very much. It's a wrap.

Wow, thank you so much. It's amazing to hear everything that's happening within the UN, be it IMF or UNESCO, FAO or UNRWA, or even Office of ICT. We have, in addition, a community of practice that brings all of us, all the UN agencies together, so please feel free to engage. We have 6, 7 members of the community in the room if you want to talk about anything. But one more time, please, you can do better than that. Big round of applause for our panelists. So we have now, we're about to start the other panel about— I forgot. Open source across level of governance, and we're going to kick the panel by a video message by Miss Veronica Gaffey. She's the Director General of DIGIT, the European Union Director General of IT.

Thank you, Under-Secretary General and Envoy on Technology. Thank you, dear Amandeep, for inviting me to open this panel. It's a pleasure to return to what has evolved in just a few years from the OSPOs for Good Symposium into a fully fledged UN Open Source Week, bringing together participants from across the world. That evolution tells us something important. Open source is now central to the digital policy agenda. Cities, regions, national governments, and international organizations increasingly face the same question: how do we remain in control of the digital systems on which our public services, economies, and democracies depend? For us at the European Commission, this question is at the heart of technological sovereignty. Not sovereignty through isolation or closed systems. Quite the opposite. It is about having meaningful choices cooperating from a position of confidence, and ensuring that our digital assets remain secure, resilient, interoperable, and aligned with our democratic values. This is why the European Commission has proposed its Tech Sovereignty package, including the EU Open Digital Ecosystem strategy. That's our new open source strategy. Today, I would like to focus on one key objective: objective: promoting open and interoperable digital ecosystems for public administrations as a foundation for strategic independence across all layers of governance. For the European Commission, we have developed a practical framework that others may wish to draw on. Digital sovereignty cannot be achieved through a single technology choice or one-off decision. It must be embedded in how public administrations design, procure, operate, and maintain digital systems. Our framework rests on three pillars: trusted digital assets that are secure, reliable, and reusable; capabilities and communities to sustain them over time; and governance that keeps technology choices aligned with long-term public interests. Together, these pillars support digital ecosystems that are both technically sound and institutionally sustainable. But open digital ecosystems are not only about software. They are also about capacity, stewardship, and avoiding future dependencies through today's decisions. Public authorities should not only be users of technology, They should also be contributors, partners, and co-creators of the systems on which they depend. This transition requires trusted partnerships and collaboration across borders. In Europe, we are already putting these foundations in place. My services facilitate a growing EU Ospol network, bringing together public administrations to exchange practices align approaches, and strengthen capacity around the sustainable use and governance of open source. We are also investing in shared digital assets. The EU Open Source Solutions Catalogue helps administrations discover and reuse public sector open source solutions, reducing duplication and accelerating innovation. New forms of cooperation are also emerging, Through initiatives such as the EDICS, that's the European Digital Infrastructure Consortium, we have the Digital Commons EDIC and the Impact EDIC, where member states are pooling resources and expertise to jointly develop interoperable digital capabilities and strengthen collective resilience. This ecosystem approach also connects to the Commission's broader technological sovereignty agenda. Through initiatives such as the forthcoming Cloud and AI Development Act, or CAADA, we aim to strengthen resilience in strategic digital infrastructure while preserving meaningful choices and reducing excessive dependencies. The Commission is also seeking to lead by example. Earlier this year, we introduced sovereignty requirements into a major sovereign cloud procurement for EU institutions. And we developed a sovereign cloud framework to help organizations assess and strengthen digital resilience. Europe's strength in the digital age does not lie in going alone. It lies in collective action. And this spirit of cooperation extends well beyond Europe's borders. Through the UN ITU Open Source Ecosystem Enabler Initiative, supported by the European Union, We are helping to establish open source program offices beyond Europe, including in Trinidad and Tobago, where early successes already demonstrate the value of open collaboration in strengthening digital capabilities. Let me conclude with one final thought. Strategic independence is not a challenge for one institution, one level of government, or one region alone. It is a shared responsibility across the public sector, from cities and regions to national governments and international organizations. The technologies on which we depend are becoming more interconnected and essential to our societies. This makes cooperation not simply desirable, but necessary. Strategic independence should therefore not be understood as standing apart, It is the ability to cooperate from a position of confidence, with meaningful choices and shared stewardship over the digital foundations on which our public services, economies and democracies increasingly depend. Thank you very much.

That was a video message from Ms. Veronica Gaffey, who couldn't be with us here today. Today, but insisted to have her message to address to the room. Our next keynote speaker would be Mr. Keith Bergelt. He's the Chief Executive Officer of Open Innovation Network, and after that we start our last panel. Keith.

Thanks very much. It's a pleasure to be here, an honor actually. It's, uh, as I was thinking about what I wanted to say, I wanted to share a little bit of my reflections on what we've heard a little bit during the week, and also just take stock on where we've come from. 40 years on since Richard Stallman founded GNU/Linux, we are in the throes of a movement that sometimes we don't think about because we're just actively participating. This movement is one that is really something much more profound than I think even many people, when they talk about open source, really can appreciate, at least in my view. And I'm a guardian of open source, and so it's something that I think about all the time. I want to make sure that people can adopt open source technology because when OIN was founded 21 years ago, there was a real threat that was posed by large companies that own significant patent portfolios, and we wanted to be able to reduce that risk and encourage more companies to feel comfortable adopting open source technology. And so in this guardian role, I think about kind of where we come from, where we're going, and this notion of taking stock, to me this is a social movement. It just happens to produce technology, which is why it's irreversible. There's an inevitability about participation, whether you are a country, whether you are a company, public sector, private sector, there's a movement that's inexorable toward collaborative development, which is one of the critical components and tenets of open source. We collaborate to compete more effectively. The concept of coopetition was— there are many, many articles in the Harvard Business Review, MIT Journal, and books written on the idea of coopetition during the '90s. Never has coopetition found its home more so than in the arena of open source. This social movement embraces the idea that we cooperate We collaborate, and we produce something that we couldn't produce independently. And we can only do it because we come together, we collaborate, and we think about how we solve problems in software together. We don't think about— the first thing we think about is not to solve it independently, but we think about what are the ways that I can build on someone else's ideas that have already been established. And what it takes is a certain level of humility. And this is something that all of us need to practice in the community. We always need to step back and think about how we can be better by distilling the collective intelligence of a population of people around the world, not just around our company or around our neighborhood if we're in Silicon Valley or Route 128, or if we're in Sophi Antipolis or wherever else in the world. We are collaborating and we are Through the process of collaboration, we have engaged and incited a group of people, the Open Source Program Office Directors, that have carried the message, have done so many different things to be able to advance the message inside companies, inside governments, to be able to awaken people to the notion of value and how value is created in the new economy. We are— I think the concept was developed— Brian Arthur, an economist, commented on Ken Arrow's work on increasing returns about 18 years ago in an article that stuck with me. And when I got into open source 19 years ago, it made me kind of think about the whole idea of 1 1 1 doesn't equal 3. It equals 6 or 10 or 20 when we bring smart people together and allow them to solve problems in software. This is a critical component of value that every day we are involved in. Every day we see people communicating with each other. CEOs have no idea who their team members are communicating with, how they're communicating, what they're communicating. They only know that by having this unbridled capacity to collaborate, we are doing things that we couldn't do and we are innovating. What the— there are lots of articles that have been written over the years, but Clayton Christensen probably at Harvard was one of the best oracles of this message that we bring the outside in to be able to enrich our environments and our organizations so that we can bring new ideas in. We see time and time again acquisitions of new companies don't get it done. Joint ventures don't get it done. Multi-party collaborations don't get it done. But open source does get it done. It increases the level of innovation that we are able to, to, to access, and it changes the way we create value in the new economy. And OSPOs are a key component that people— I've seen them evolve over these 19 years, and I've seen The Japanese team that was up was talking about— they're the next generation of OSPOs, because I can remember when Fujitsu founded the first open source program office director position in Japan, probably about 20 years ago. It was right before I arrived on the scene, and I thought about how this role has evolved in preparation for these making these preliminary comments, and I see that the increased kind of importance of these individuals and the role that they play, with humility— they're not the, you know, just like I think, you know, Jim Zemlin spoke to you earlier, Mike Malinkovich will speak to you in a little while on the panel. These are people who are doing really significant things to be able to create what I'll call project-based innovation, which is the most powerful force for innovation in the world right now. That's not talked about, but that's how we innovate. We get together in projects, we have talented people that are managing these projects that we bring in, that develop processes and procedures to be able to allow people to come together and to be able to create something that would not be available or accessible without this notion of collaboration that's so central. And so you think about the idea of their— Ospos are translators. They take a message and they translate it. They help provide kind of a conversion of raw ideas and concepts and make it saleable, make it palatable to individuals that are potentially leaders of companies, CTOs, deputy CTOs, heads of software, and other individuals, not only in their own companies, but across the industries that they participate in. They're oracles in their own way of the message of how powerful open source is. And then, you know, they're communicators. They provide a vision, a set of norms, a code of conduct around how we participate in a community which is that has the twin components of opportunity and obligation. We have an obligation to participate in a certain way. We have an obligation to lift each other up. It's not Fujitsu competing against Hitachi. It's about both companies raising each other up with processes and procedures that help us to be able to redefine what innovation is and to accelerate the capacity for innovation that only open source unlocks. They're advocates. Very often they've had to fight inside companies, particularly early on in the history of open source, to be able to have people become aware of the importance of this movement that is open source. Many people in senior positions inside companies get to some extent of it, but they don't get it all. They don't truly understand the power of open source as I'm describing it and as I understand it. They're educators. They're not just communicating a message and establishing norms and a sense of who we are and how we participate and what the expectations are in this community. They're educating people all the time. They're creating a sense of value, and they're also doing things that other people can't do. There's a special Again, the special sense of learnership that I find and I found from the beginning in these individuals who are connectors. They provide a bridge between the outside project communities and the inside parts of the organization. And it's all parts of the organization that they touch. I think Brian Proffitt wrote a great piece that's on the website on why— what is an OSPO, how do they operate, and I think it's a really instructive piece on— that has you see and understand what OSPOs do on a daily basis, how they contribute, and the incredible value that they have. And OSPOs in the private sector, I believe, as we head into this panel, provide an invaluable role and a source of knowledge and awareness that is invaluable for the public sector to adopt and understand so that we can all be singing from the same hymn sheet. We can all be understanding the power of open source as a social movement that produces innovation that is of incredible significance and import to all of us in every sector and every industry where we find software. And so with that, I'll also mention that next week I'll be announcing in Japan on the 30th that we will be starting a special initiative to work with OSPOS to allow them to participate in an extension of their incredibly important role in companies by having them understand through some software that we've developed where the gaps are, where we should be providing more protection and coverage in their companies. By— as they— they're auditing on a regular basis how much code is being used, what licenses they're done under. Their— the compliance and governance role that they perform is invaluable, often working with lawyers inside their companies. But what they're doing is now going to be extended to allow them to identify the gaps. Where are they using technology that we, OIN, are not protecting? And then nominating that so that we can protect fully the scope of open source technology that's being used. And so I could not do this with just people from the Office of General Counsel or from the IP departments. I need to use people who have this unique access across the company and a unique understanding as true believers. And that's— these are the people that I rely on. These are the people that are my greatest advocates. In companies all around the world, not just in Japan where I've spent years as a diplomat, but everywhere in the world. And so this is a great resource. This resource needs to be utilized now to be able to ensure that we are spreading this common sense of knowledge awareness and distilling this collective intelligence and sharing it. Thank you very much.

Wow, thank you so much, Keith. It's incredible to see that, to go back through the history of all these human interactions that are involved for us to be able to reach a tool like Linux is literally millions of human interactions, not code, forget about the code, is how many, the size of a community and ones-on-ones that took place place for us to be able to use those digital goods that are changing everybody's life. With that, I'm happy to call for the last panel of the day. Last panel of the day. Please bear with us. This is the Strategic Independence Across Layers of Governance. I'm going to call on our panelists and please feel free to come in. We have Anne-Catherine Riedel. She's the Managing Director of Next in Germany. She will moderate our panel today. We have Dr. Indra Pradana Siguinatas, he's the Secretary-General of the Asia Productivity Organization. Please join us, sir. We have Yun Barker, Yun Choi Barker, she's the Deputy Director-General of Information Technology Department at the Asian Development Bank. We have with us, of course, Mr. Mike Milankovitch, Executive Director of the Eclipse Foundation. Nitya, Nitya Ruff, she's the chair of the Linux Foundation board. And then Brian, Brian Behlendorf, independent advisor and board director.

Please.

Thank you, Omar. It's a great honor for me to moderate the final panel of this conference today. And without further ado, I'd like to directly start and introduce my panelists with their first question. So we will talk about strategic independence across layers of governance. And as Omar said, I'm from Germany. It's a big discussion in my country. And as Ms. Gaffey said, also in the European Union currently. So I'm happy that we will discuss this panel here with leaders from international organizations and industry leaders. And I'd like to start with you, Dr. Indrapardhana Singhawinata. You are the 13th Secretary-General of the Asian Productivity Organization and in your second term. And one of your tasks is to, in your position, to encourage your member states to use AI to boost productivity and help them build capacity in emerging industries. What role does open source play in this work, and what does strategic independence mean for the work with the countries you're working with in general?

Thank you, Anne. Good afternoon, ladies and gentlemen, colleagues, all the experts. I'm Indra Singha Winata. APO is an Asian Productivity Organization established in 1961. Now we consist of 21 member economies from Asia, from the advanced country Japan, emerging ASEAN, and LDCs. Jump into the answer to your question, Anne. So at the moment, yes, starting 2022, actually APO, by the request and the demanding coming from our governing body in 21 members, They would like to transform our methodology of the productivity that in the past, after 6 decades, we are doing the productivity methodology, productivity training in the manual and basic. Now we would like to transform in AI. We have a terminology coming from Anne. First is the open source itself. Frankly speaking, it's very help of our member economies. And by the data of APO, it already can show that by having the open sources AI implementation, of course, this is not from the whole member countries because we have 21 with the various level of economies, at least for the emerging countries, who are very hungry and thirsty of this technology. We can already can prove these 3 years by having our demo company projects and our capacity building projects. We can have the result, and the result is reducing the cost quite significantly. Now we talk about strategic independence. With the open source also is very help, very much help to the members government, because by having this open sources now, the government in each of our member countries, they can choose. Now they have an option, not only one or two big players, but they have an option to choose the open sources system, the AI system, that match, fit with the needs of their development countries. By having said that, to adopt the tools is quite easy, but actually to have the ecosystem that I mentioned last Monday, the ecosystem is very, very much important. It's a stage of end-to-end awareness, adoption, skills upgraded, and finally come out with the regulatory framework. Because innovation— I'm in APO almost 4 years in APO, I can see how the gaps, how the disparity within our members is still there, which is okay because all the states have their own states of development. But what I'm trying to say is two things. When they adopt, it is very good for them if they can follow up with two things: the workers' skills ability upgrading, and second, follow with regulatory framework. Thank you.

Thank you, Dr. Indrawati. And I'd like to introduce to you Yoon Choi Barker. She's the Deputy Director General of Information Technology of Information Technology Department of the Asian Development Bank, and you're based in Manila, but you worked before here at the UN and supported the UN Secretary-General's UN 2.0 vision with your work here, among other things. Also to you, what does strategic independence mean for the work with the countries the ADB works with, and what does open source play for a role?

Great. Thank you so much, Anne-Catherine, and Good afternoon to all of you who's been— who's staying with us until the very last session. Thank you so much for that, and I'm very honored to be in this very distinguished panelist for this important session. So as you said, I have recently moved to Asian Development Bank, and from our perspective, strategic independence does not mean countries necessarily own or build their own technology and all the various technologies. That they use to provide public— provide services to their citizens. For us, it means really enabling the government to set their digital priorities, making informed decisions, and being able to ensure that they are not locked into a specific vendor or specific platforms that may be difficult to get out later on, and being able to have that flexibility and the freedom to do what is best for their overall priorities and strategy of the government. And this also is very important to our developing countries, as Dr. Indra just mentioned, because now more and more all our public services are really depending on the digital systems. And especially artificial intelligence has become what everybody is leveraging and having the hunger and the desire to incorporate in all the various services they deliver and various social matters, whether it be climate or healthcare, education, and so on and so forth.

So the—

so we believe that open source, with its transparency, reusability, local adaptation, and continuous improvement, and that the various communities and all the different groups provide to continue to keep it up to date and even more robust potentially than some of the other vendor-provided platform, this is really a great opportunity for many countries, especially developing countries, who may not have the economic means to be able to purchase those products and platforms that are provided by some of the few global companies that can potentially be locked into. This is where The open source is something that we continue to need to encourage for its transparency, reusability, and so on and so forth. So this is also where the multi-development banks such as ADB comes into play. And we not only provide a support around helping the government to choose the right technology, and open source, of course, is one of those as well, but making sure that they are equipped equipped to make the right decisions and their overall strategies aligned to the platform choices that they make. So open source by itself is not a means to, you know, to the— it's not an end to it by itself as a technology solution, but really it should be linked to their broader digital strategy, digital direction of the government and the various countries. In that context, that we really want to not prescribe specifically as we work with different member countries around what they need to do, but really more of equipping them and providing an enabling environment and ecosystem around capability building, around providing various options for them to pick from, as well as the full lifecycle of managing digital solutions and platforms. It's not just about the launch itself or leveraging a specific software or specific datasets, but about how to continue to evolve and be flexible around leveraging the insights on the different platforms and the datasets that they're using to ensure that the government is making the right decision for the public interest and for continuing to support the sustainable and inclusive and resilient development impact.

Thank you.

Thank you so much, Jun. I'd like to bring in Brian Bilendorf to the discussion. You're an independent advisor and a nerd diplomat, as you call yourself. And I just learned you co-founded the Apache web server project and gave it its name. Very interesting story. And you switched through the sectors during your career and have therefore an inside and outside perspective on government. So in your point of view, what should governments focus on regarding strategic independence and what has changed in the discussions over the years?

Thank you. Yeah, I've been fortunate enough to have played early and/or significant roles in a number of different open source foundations, in Apache, in the Linux Foundation, and in the Mozilla Foundation. And all three of those are very different in their theory of change and the way the way that they operate, and in their sustainability models. And recurring throughout all three of those has been this desire to— and a desire on my part to think about longevity. Like, how do we build not just the technical infrastructure that societies can reliably build upon and not worry about the rug being pulled out from underneath them, you know, the following year, or for all the support infrastructure on want that to disappear. How can they know how to participate in the road mapping of that and know how to depend upon it going forward? And I think about governments engaging in open source and building lots of open source technologies, and I worry. I worry that any one of the member states of the UN is one bad election away from a chainsaw-wielding rocket man walking in and deciding that that Everything that had been built inside the government was no longer necessary, right? And so, I've also played roles at the World Economic Forum. I worked in the Obama White House just before the USDS and 18F were set up, which are fantastic— were fantastic things to stand up to try to build in-house capacity inside the U.S. government for this. And I've seen them inspire digital service teams across lots of different countries inside of lots of different states. And I think those have been critical to demonstrate a new way of engaging with technology, that governments can engage with technology and try to up-level the engagement with the private sector. But I think it's really critical that government view a very— determine a very productive relationship to have with private sector actors who can be part of delivering that last mile of the solution. In fact, if we characterize this type of organization as a digital delivery services firm, that could be an outside provider of that solution. It could be a small company. It could be a large systems integrator, or it could be an in-house team. But they play this pivot role, right? They're pivoting between delivering the value for a government customer to the citizens or other stakeholders on one side, but also pivoting upstream and working with an open-source foundation or just an open-source project to try to make sure that not only that they're getting the value, but that they're also contributing to the long-term health of that project. And I, I think key to figuring out then, uh, to get back to the question, uh, the, the, the ability for countries to to be self-sufficient in their technology, to not be beholden to vendor capture, not be beholden to proprietary software. The key is figuring out how to enable the kinds of vendors who are going to do the right thing, the kinds of digital delivery service teams, whether in-house or outside, uh, and giving them an advantage in the procurement process, in the technology selection process, uh, and trying to really systematize that relationship so that as a country, no matter what no matter what size you are, large or small, you have optionality when it comes to supporting the digital public goods that you're deploying inside of your organization. Lots more to say, but I'll pause there.

Thank you, Brian. And I'd like to welcome in our round Anithya Raghavan. She's the chair of the Linux Foundation board and one of the world's most recognized open source leaders. You worked in the industry and now a lot with universities. And the same question to you: what should governments focus on in terms of strategic independence? And maybe, as you are also a pioneer of the OSPO movement, maybe you can focus also on that in your answer. Thank you.

Thank you, Anne-Catherine, and thank you all for staying to listen to this very important discussion. I think there have been a lot of shifts that have happened over the last 20, 30 years, particularly in the last 10 years. Open source for governments used to be more about cost, more about procurement. And today it's about strategic infrastructure, not just a procurement category. But, uh, having witnessed some of the fragile supply chain issues, uh, during the pandemic, uh, looking at geopolitical fragility that we, um, experience today, and also tensions, the acceleration of AI, all of these have resulted in digital sovereignty being a public policy issue, not just a technology preference. And I think there's another shift has happened in the 10 years or more ago, people used to just consume open source, but the most important shift that I'm seeing happening is people are saying, I need to contribute, I need to be at the table, I need to steward open source, I need to not just be a consumer. I love the example of Germany's sovereign tech fund. The European Commission has also taken a stance on owning open source and, and contributing to open source. When your staff are upstream contributors, they know how the code works, they have influence in the communities that shape it, and they have genuine options when things things go wrong. I think reforming procurement has also been another shift that I have seen governments make, and I think that's a very, very positive shift that's happening to say open first. Until procurement changes, open source always has remained marginal in the past. The last couple of things I'll say is I've seen a shift in building internal capacity, and we've heard from so many ministers in the past 2, 3 days that they're actually building capacity to be engaged. Yes, it's unglamorous, but it's a foundational capacity that you need. Strategic independence does not come from just consuming or saying that you have a policy. If you— in complete technical workforce, our contractors, when they leave, your knowledge leaves, and you don't have independence. You have— continue to have a dependence. And I'm glad to see some of the regulations that are also happening, the CRA and the AI Act and the interoperability requirements. That creates a policy environment that actually supports independence in ways that did not exist in the past. So I'm seeing a lot of great changes that are happening, and Katherine from a government perspective. And I think OSPOs can serve really as a function to direct some of these changes and to connect all of the institutions across government so that we don't have silos in a lot of these different areas, and to really connect it from procurement to legal to business to external community. So those are some of the things that I would say government need to do.

Thank you so much, Nithya. And I'd like to welcome Mike Milinkovitch. You're the Executive Director of the Eclipse Foundation, and you are, of course, also recognized industry leader and open source community champion. And this morning, Adriana Grosz said that the open source community literally run the world. Do you agree with her and what is your take on the open source community and their strategic role— their role on strategic independence that governments strive for?

That's a lot to unpack. Well, first of all, thank you for having me. I really appreciate it. For those who don't— have never heard of the Eclipse Foundation, we're the— we're a 22-year-old organization 5 years ago, we moved our domicile from the United States to Brussels, so we are now the largest open source foundation in Europe and host quite a few projects, many of which are related to European regulation and the search for European strategic independence. I think that the role of the communities play is vast, because obviously the code that runs all of these systems mechanisms are incredibly important. And I just— but I've just recently read a quote. Some of you might have heard of this. It's malus.sh, M-A-L-U-S. It's like a— it's sort of a parody, bit of a parody website that has a mechanism by which you can reverse engineer open source projects using AI and basically makes the claim, so we don't need open source anymore. And it had a quote in there that I really find really fun sort of matches on this question, which is that the global economy now runs on open source code which has been written by— written by— sorry, now runs on open source code which is technically owned by no one, which is maintained by developers who are technically employed by no one, governed by licenses which are technically read by no one, which is sort of true, only sort of true. It is a miracle of human cooperation. Also, from a fiduciary standpoint, completely insane. And I just want to plant that seed, because the way that the world works, and I really strongly believe that open source, I'm sorry, it runs the world, it really does. No one could have ever designed this system that we have today. It has evolved through steps by things like Brian did many years ago and all the evolution. And then we've now got the attention of governments as communities. And Nitya mentioned the Cyber Resilience Act. I mean, up until 3 years ago when that was first being drafted, I was— I think I would have said somewhat proudly that I had never been involved in a policy discussion. Then the first draft of the CRA came out, and I would call that, from my perspective, a near-death experience. It was. It would have been horrific. It really, honestly, it would have been. And it was entirely with good intentions, but we had to really buckle down and engage with policymakers in ways that we had never done before. And I think that that really sort of sets the stage. What you see now from open source communities and open source foundations is a much stronger sustained investment in dealing with policymakers, dealing with governments, and I think that that ultimately is a very positive outcome from what could have been a very negative event. So I'm very optimistic about the future. I think that there's been more conversations between governments and societies and, and open-source foundations and communities than there ever have been in the past. So yes, I remain relentlessly optimistic that this is only going to get better.

Wonderful, thank you. And as we're running out of time, I'd like to ask one last question to the whole panel. As governments increasingly rely on AI, what are the new dimensions of strategic independence that policymakers should be considering? Who likes to start? Whoever wants.

Go.

I'd be happy to comment on a couple of things and then pass the baton. I think a lot of people have talked about data, getting your data house in order, break down the silos in government data, make sure you understand the provenance of the data, and do your due diligence on the models that you use and the data that it uses. And one last thing I'll say, and then I'll pass it on to others. I think governments need to have internal capacity to evaluate AI systems independently, to test real-world performance, assess the risk, and also challenge vendor claims. Without this, informed procurement is almost impossible.

I think a lot of the conversation around AI has been reduced to conversations about the large language models, when actually a full AI system is a lot of things built both underneath the LLM but also well above that, at the layer of harnesses, at the layer of apps that talk to the LLMs, those sorts of things. And in thinking about strategically what's important for governments to— and, and enterprises of all sorts— to be able to intervene and modify behavior and not control but, but, but have agency over and direct. I think it's a lot more important that at the harness layer, at the layer that wraps around an LLM, that we have capacity and we have open-source tooling and we have institutions that know how to manage the wild and wacky stuff that might come out of the black box called the LLM.

Thank you. Yoon.

Thank you. So, we had an earlier session on AI for digital public goods where we looked at the The fact that we, so far, from the Digital Public Goods Alliance perspective, not a single product yet, AI system, has been certified yet because it's a complex discussion, a complex issue. It's not just a software that you just install and start running with it and evolve and enhance. There are many aspects and facets around this, especially around the data, the safeguarding of sensitive information of public, of the citizens and so and so forth, and the continuous evolving of the model and its weights, as well as contextual as to where it's being trained and how it's being localized in the various places. So just because it works in one country does not mean it's going to automatically work in another country. So this is really a complex discussion. But however, again, with open source, I think this is where the freedom of independence comes in, where the localization is so much easier to leverage or to implement. And to be able to continue to leverage and evolve it, leveraging a larger community of people that are supporting the various platforms around the AI and not being, again, locked into a specific product company or a specific country or vendor that the product is being developed from. And most important, I think, is the accountability. The decision that gets made by the AI at the end of the day really is owned by the government itself. It's not the product or technology that makes the decision, but it's really the government who needs to own it. So make sure that the government has the institutional capability to monitor and intervene as needed and be able to challenge the decision the AI is making and being able to really responsibly using the AI. And this is is, you know, goes back to the responsible use of AI, is such a critical, important aspect of how we're leveraging this and giving more independence to the various governments who's adapting it and really leveraging it well. Thank you.

Um, I— when it comes to how AI is going to impact government, I, I want to actually take a very different and quite specific point of view, um, which is around the topic of security, uh, cybersecurity specifically. So governments all have digital services. They have basically— governments have basically become the IT department of the nation. Prepare to get your ass kicked. The— there's— more than one thing can be true at the same time. So Project Glasswing was an excellent marketing campaign by Anthropic. It was really well executed. It's also true that, um, Mythos is a very powerful model that delivers, um, lots of vulnerabilities. More worryingly, um, what it does is it actually delivers a playbook, an attack playbook. So it's not just, "Here's your vulnerability." It's like, "And here's how you exploit it and here's the script." Um, it's going to be messy, uh, and every IT department, uh, of which governments are one, need to fundamentally change the way that they think about their running systems. If you are not prepared to repatch and reboot every server in your infrastructure several times a day, you're going to be in trouble. And so, and that's a huge cultural change. Every IT— I mean, I've been in IT. Every IT department in the planet, like, you— and, you know, you put it up, you stand it up if it's work, If it ain't broke, don't fix it. Right? And they just run this. I remember talking to like one of my own IT guys, like bragging about the fact that he hadn't rebooted a server in 9 years. Hmm, excellent. But that's, and that has been the perception of IT departments for as long as there have been IT departments. And fundamentally that's going to change. And I think that we really have to think as governments, you have to start really thinking about how you are going to change the culture and the security posture in your IT departments to be relentlessly using AI on the defense. It's not because then there— because there are definitely ways to do that. Again, to Brian's point, you have to have the right harnesses, you have to know how to use the right prompts, you have to train your people to do this, but you need to start really using AI to automate your cyber defense.

Thank you.

Dr. Indra.

Thank you. It is very true that actually is no longer about how to focus on the digital infrastructure and software itself, but based on the survey that and study that APO have with our 21 member countries, the most important is have several aspects. One is Government data governance. Second, government transparency, government accountability, even the government human resources itself, that we should be really know about this new technology. And lastly is about how to manage the algorithmic risk. That's also very important because If the government uses the AI system that finally cannot understand, cannot be audited, cannot be explained, then actually the strategic independence is fragile. It's nothing, actually. This is based on the study. And finally, that what I would like to share, that AI sovereignty without the skills and the governance capacity inside the government itself, if we want to— if the government really want to do this AI and open sources thing, it's become not really sovereignty. That's my last statement. Thank you, Anne.

Thank you very much. And as Omar asked me to end this panel 7 minutes before 6, and we're now 8 minutes— no, now 7 minutes on time. I'd like to thank you, my distinguished panelists. A big applause you. Thank you for your perspectives.

And I can thank you, Mike. And indeed, the software open source runs the world. And thanks a lot for Adriana. And we did it on purpose to leave our best speakers at the last to make sure that you stay all the way till the end. So please Join me for a big round of applause. You can do better than that. This is it. Looks like you survived the osu! for good day. Congratulations to you as well. I hope you found the day very interesting. For the send-off, for the final, final speaker and to close the 4 days that we had at the UN, as you know, Open Source Week is not over. There will be a slide at the end where we're going to show what's happening tomorrow. A lot of things are happening tomorrow. Our team has been working very hard, led by a program committee led by Sachiko, Brian, and others. So we have 20— no, 30 fantastic sessions for you tomorrow. That's going to be offsite. That's our tradition, that the last day is not— we don't organize. We give a space to the community to organize their own event. But before that, I want to give the floor to His Excellency Mr. Amandeep Singh Gill and the Secretary-General and Special Envoy for Digital Emerging Technologies.

Thank you.

Thank you, Omar. Thank you so much. And I'll be very brief, just a few words of thanks and some impressions I have from this exciting, wonderful week. First, the thanks, the co-hosts, our partners who helped us organized this week, all the sponsors, the volunteers, they've done an amazing job. Please, a round of applause for them. And you, all participants, you know, you have been amazing, so thank you. Now, it's very hard to recap the week, and I won't even attempt it, so it's just impressions. I think we experimented this year with a full day devoted to Open-AI, and we realized, and we heard some of the practitioners, that we are on a spectrum. It's a moving dot. It sits somewhere depending on context, but it's moving. The concept is evolving, and it's important to have this discussion. Now, sometimes this can be a contentious discussion, but it's important that UN remain the space where even contested issues can be discussed constructively. We also had from that day a clear takeaway. Yan Lekun mentioned it, that excess doesn't equal capability, and actually capacity is the bridge between where technology sits and where actual outcomes are realized. So not just Yan Lekun, but ministers from Jamaica, from other parts of the world echoed that. So, I think if we were to just capture this in one sentence, that particular day, I think the sense was that we have to kind of balance between rented permission and own capability, find the sweet spot depending on context. So, that is where agency lies, that's where sovereignty lies. So this is more than adoption, perhaps it's even more than adaptation. So this is adaptation plus, but we have to find our own sweet spot. And then fragmentation is the common adversary here, and shared inclusive infrastructure is the response, and that's where open source becomes such a powerful tool. And then, you know, the digital public infrastructure day, which is now a regular feature. I think this year we saw some very powerful regional perspectives come together from the Caribbean, from Maghreb, from Central Asia, and from, of course, Europe. And it was interesting, in speaking to the ministers later, they really appreciated the opportunity to learn from each other. The deep dive we did into financing of DPIs, also with some philanthropic foundations, Co-Develop, Action, the multilateral development banks, World Bank, IDB, that helped us make it more concrete for the ministers, the vice ministers that participated. There was an honest discussion of gaps and challenges, not just building DPI, but how do we fund it, how do we sustain it. And one of the things that really resonated with me was that we have to find— we have to own it. So, skin in the game. So, national budgets, even if initial bootstrapping comes from the outside, national budgets have to have lines for DPIs. These are the digital roads and rails on which everyone can travel to the future. So we must ensure that, you know, we are maintaining them, fixing the potholes, ensuring reliability, and keeping it safe and accessible for all users. Now, coming to this day, the Ausposts for Good, really the heart of the week, I think my impression is that the community stays the center peace. So this was always our first objective, how to bring the community together, make it stronger, make it more diverse, and make it more, in a sense, relevant for the time we are living. So we saw that highly engaged room, packed room, so growing momentum behind open and collaborative approaches. Many of And the projects and initiatives you heard about, you know, not only from the UN but also outside, you know, they demonstrate that. So the community is ready to go from these positive vibes, these principles, to actual action. So there were some nice moments for us, you know, the report on AI as a digital public good, the launch of the open source enablement compass to track how the member states are doing in their open source adoption journey. Some markers, in fact, as we go ahead, and we heard from directors from different UN entities, UNESCO, FAO, UNRWA, how this is so important from the perspective of the reform of the organization, the UNAT process, where we deliver better, deliver stronger, deliver together, with some challenges on resources. Now, there was one last element that I wanted to say. Community staying strong, community coming together, translating principles into practice, but then how do we go from here to next year? I think this is where your valuable feedback will help us. We've done some things hopefully better than last year this, this time around, but next year we want to make some more improvements, perhaps look at, you know, how this community can orient, reorient in the direction of some of the challenges that the AI turn faces— data availability of high quality and diverse data. So how could data commons be one of the additional responsibilities of the open source community apart from the code side, the model side? And then there are values that we need to rally behind, that we need to take forward. There is some interest politically these days in digital sovereignty. Open Source is having its moment in the sun in that context. You know, sometimes political trends, you know, they, they evolve, they get on to the next challenge. So how do we stay true to our values? How do we ride these political waves up and down? I think those would be important considerations as we plan for the next Open Source Week. So you'll soon get to the the GitHub reception, and then tomorrow is the community day where I think more excitement is in store. So thank you very much and look forward to seeing you next year.

Again, thank you very much, Under-Secretary-General, for closing and summarizing all what happened this week. It's been very, very intense 4 days full of learning, full of panels and keynotes and fireside chats, and most importantly, hallway talks and lunches and coffees. I hope everybody managed to get the best out of the conference, expanded their network, met amazing people. I'm sure, I hope a lot of new initiatives are all going to start from here. And with that, I am happy to close the day, but before that, we have one last slide. I promise this is it. We have one last slide about what's happening tomorrow. Tomorrow, as you know, it's not happening here. This is the tradition that we don't host the last day. The last day is your day. It's the community who runs it. And our incredible program committee led by Sachiko have done a lot of work. They did the call and something like 250 sessions were submitted from 50 countries. They sorted through, they read through every abstract, everything, and then they made sure to sort only the very best. They combined, they concentrated. So now we have those 20+ sessions. 110 speakers. So literally, the community is talking to the community. We have no business with that day. That's a day that is run by you, run for you, not even hosted here. It's hosted outside of premise. So thanks to Red Hat who are hosting us over there. We look forward to seeing all of you there tomorrow. And just to close, just wanna thank the team who has been wonderful, putting everything together. It's been incredible work. As every event, there were 120 dramas in the background, but a good event is an event where nobody noticed the drama. So I'm very happy about that, but we would never have been done— we would never be able to do it without my colleagues. So Zeyna, Mitusha, Moritz, of course, all our partners, Francesco, Toby, Brian, Caroline. There's just— you— oh, of course, of course, our— there's just so many. It took an amazing collective effort to get to this. And I think with this, I see you next year. And I promise it's not going to be Open Source Month. It's— we're starting our Open Source Week.