15th UN Crime Congress: Committee I, Workshop 3 – 29 September 2026

Good morning, distinguished delegates, ladies and gentlemen. Welcome back to workshop three entitled, "Workshop on Getting Ahead: Strengthening Data Collection and Analysis to Better Protect People and Planet in Times of New Emerging and Evolving Forms of Crime." This is the third session of the committee one of the third workshop and I would like to welcome our moderators Mr. Matti Joutsen and Mrs. Letizia Ferhati and yeah and with this I'd like to give the floor without further ado to Mr. Joutsen to start the first session of today's meeting the floor is yours

Thank you very much, Mr. Chairman. Good morning to all. It's a pleasure to moderate this second session. Yesterday, we had a very good, I would say, robust session on data. Basically, what we did yesterday was look at the different layers of data that have been used in trying to analyze crime, understanding crime, and then providing information for decision makers. And the way that I looked at how the Thailand Institute of Justice has structured that first half, there were layers. It began with the information that is produced through the operation of the criminal justice system police contact, court action, correctional system. Then that was expanded to the perspective of the victims, but that was within the self-contained structure of the criminal justice system. Then added to that was a second layer, which was looking at the data that is produced by different government agencies, which is needed when you want to look at other types of issues that are new and emerging that hadn't been recognized before as traditional, if I may use that term, forms of crime. We're talking, for example, of money laundering or undocumented migration, et cetera. And there you have other agencies that are on the first line of responding and that need to cooperate with the criminal justice system. From there, we went on to for the private sector, which becomes very important when you start talking, for example, about money laundering, but also about fraud, computer crime, and other forms of new, evolving, and emerging crime. And further layers were added by academia, civil society, et cetera. My takeaway from the discussion yesterday was There's a lot of data. Data exists, but it's fragmented. It's in different formats. It's held by different entities, different people, and they use different indicators. And one of the issues that we have now is trying to somehow manage this data, get access to this data, and use this to be able to better respond to not only the traditional forms of crime but also to the new evolving and emerging forms of crime. And those new evolving and emerging forms of crime are becoming more and more of a problem. I mentioned yesterday when I was responding from the audience that five years ago I was involved in a study done by UNICRI and the Thailand Institute of Justice on threats within the Southeast Asian context. And I had the opportunity just a few weeks ago to look at a new UNODC study on the threat assessment of organized crime within Southeast Asia, and I was really taken aback. I was astonished by the fact that you had things like fraud centers, boiler rooms, scam centers, whatever you want to talk about, or online child abuse, forms of crime that did exist in kind of vestiges five years ago, but really weren't that much on prominent. Another example, of course, would be computer crime, which is something that almost everyone now has contact almost on a daily basis. But because those threats are real, sorry, those threats are new, they are difficult to detect on the basis of the blinkers that we have, which is based on our own, our existing sources of data. What we will try to do during the second half of the workshop is go beyond those more traditional existing forms of data and see how they could be harnessed, managed to better respond. And what we will be doing is first we'll be taking a kind of a visual scan of how do you go from the present situation to looking at what crime threats would look like in, say, five years or ten years down the road, if you don't even know what that crime would be. After that, we will be looking at data management, data security, the standards that member states have agreed to informally on, sorry, through the United Nations system on using these data and storing this data. And then after that, we will be talking about geospatial data, international data flows, and getting to what I think would be the essence of any discussion with the UN, the data context, and that is international cooperation. We're beginning to understand what we need to do on a national basis, but we also need a discussion on the international aspect of this, how do we cooperate within the United Nations system, how do we cooperate with international actors, and in general, how do we improve our not only national but international response to crime. Having said that, it's my pleasure to give the floor to our first speaker, Mr. Ari Varjok, who is a Finnish public safety expert. He's Finnish, he must be good. There you are down there. Sorry for my Finnish nuance on that one. He is presently working as the chief executive officer of the Finnish Solutions Fund Effect. He's a former police officer and intelligence professional who spent 14 years at Finland's Ministry of the Interior, most recently as its director of strategy and development. And he is here to tell us what we need to know, not about prediction in the sense that is often used in law enforcement, but about foresight, trying to look ahead. Ari, the floor is yours.

There we go. Fantastic. Esteemed Chair, Mr. Moderator, Your Excellencies, thank you for this opportunity to share some thoughts and reflections on a subject that is very dear to me. I'll be drawing from experience of a journey at the Ministry of the Interior in Finland that started around 2017. namely I'll be talking about strategic foresight in the context of crime prevention and law enforcement. Compared to yesterday, which was a very fruitful initial part of the session, I'll perhaps bring in a somewhat different beat. I don't represent a research organization or perhaps an organization that produces data, but rather government authority or perhaps reflect on my experiences from a government authority where we were the utilizers of crime data and of course lived in very much in the kind of a juxtaposition between research and the founded data and then the requirements to make decisions. And often, obviously, very quickly. Now, my main claim here today is fairly simple. Historical data on crime is useful for operational planning and academic study, but a lot of times it is insufficient for strategic planning and good policy making. This referring to the juxtaposition I just mentioned. I might be somewhat provocative, so bear with me on this. So basically, in addition to asking what happened and why, we need to be asking what could change in a way that old rules don't apply anymore. Equally, we need to understand what skills and capabilities we could need in the future, and also perhaps ultimately what our desired future is and what we need to do to get there. And these are obviously processes that happen subconsciously, unconsciously, or non-directed or directed. So it's really a choice of whether we want to be steering that process or not, and which things we identify that should impact our answers to those questions. Now, like I said, in Finland in 2017, we started the systematic foresight work as a part of the government strategy of internal security, which I had the privilege to be a part of in drafting. Our reasons for this were very pragmatic and down to earth. At that point, even our so-called strategic reports, and up until then, they had been quite reactive and threat-based. This was obviously good for operational preparedness, but in many ways politically insufficient and misleading. Firstly, they didn't take into account what would become better with time. This led our reports being read as scare tactics to get more funding, and to some extent, you could argue that they lost credibility with the political decision makers. Another element to this was that our perspective was quite a bit from within. We often failed to identify both how slow macro trends and sudden events impacted our environment in the long run. A third result of this was that we didn't challenge or revisit our strategies, methods, and partnership for combating crime in a changing landscape often enough. This led to inefficiency in many situations. For example, during rapid growth in cybercrime in the mid-2010s, We initially continued doing more of what we did before, i.e. investigate and prosecute, until we perhaps at some point came to the understanding that the overwhelming mass of cybercrime doesn't really respond quite well to this, and we're not achieving our goals with the approach we had had and taken previously. Now, as our work with Foresight evolved, we devised something of a three-step maturity model to assess where different parts of our law enforcement and crime prevention community were. The first step I already more or less described, which was operational forecasting, where we use historical data to plan operational responses like temporal and geographic hotspots. You know, in Finland, ski resorts in the north are highly populated on winter breaks and weekend nights in city centers are busy. And this obviously reflects in crime. We already had some elements of the second step of our maturity model, adapting. We identified potential changes in the near future in the form of threats that we prepared for, but we lack some other insights. Namely, most larger societal changes have both positive and negative outcomes for public safety and crime. Again, drawing from the Finnish example, something we missed to some extent, at least, was that we have a very rapidly aging population, which will almost universally decrease street crime, but at the same time make us perhaps more susceptible to internet fraud. And we both these trends to some extent. Another example from 2014, combating gas theft at gas stations was a major problem. By 2020, this problem had virtually disappeared due to gas pumps largely incorporating payment terminals. And these were changes we didn't really foresee when we were in the thick of things. Now, looking at the third and most ambitious level, perhaps shaping future, i.e. having a desired future is something that's perhaps, at least for us, was quite foreign within the context of criminal policy and crime prevention and law enforcement. We used to react to threats, whereas we needed perhaps and wanted to be proactive towards a safer, safer future. And those are There's a nuance difference between those approaches, perhaps, and we clearly lack this added layer. The central question being, what future do we want? What are our desired safety outcomes? And of course, we had strategic goals that politically mandated strategic goals that to some extent address this question, but not on a very operational level. And obviously, after having determined kind of our position on these things, we can try to answer the question, what future we want, and start extrapolating and backcasting based on distinguished previous data, research, other countries' experiences, and expert opinions, the answers to two questions. Which social drivers do we need to affect, and who are the key partners we need to engage in achieving these? Now, Perhaps to conclude in a somewhat practical illustration of what we did and are currently still is the practice. There's basically a three or four step model in the strategic strategy cycle. You have the politically set strategic goals, which you can see on the left. These are current goals that are in force. And On a yearly basis, we then look at the known drivers and components behind these goals, i.e. which things affect the accessibility of urgent services, what forms of violence are emerging, what drivers in society impact violence, so on and so forth. And then also we like to look at unknowns and obviously different scenarios. I won't be obviously going into methodologies that much due to time constraints, but kind of managing uncertainty and scenarios obviously is a way to to do this. And that's all fine and well. That produces another type of data, but you need to operationalize this. We use three tools, mainly, three main tools. The first one being the impact certainty matrix, which I'm sure is known for a lot of you, where we look at what drivers and phenomena have a high impact and high certainty for our goals. of a mapping of this and then perhaps do some prioritization based on that. But apart from that, also looking at the impact influence matrix, i.e. what drivers have a high impact and we potentially have within our branch of government or government as a whole, high influence over and subsequently perhaps look at what partnerships we need to form to be able to affect especially the ones that have a high impact on these strategic goals, but we perhaps organically don't have a high influence over, say, for example, working with a third sector business, civil society, things like this. And this really distills into the steering process a 1+4 years time frame where we set concrete performance targets for the agencies. You could say it's a continuous one plus four year steering time frame. It's adjusted annually, but constantly looks five years into the future and trying to be future sensitive, perhaps, within the regular year to year steering cycle. This is obviously an attempt to harmonise immediate next year concerns with then evolving long term planning needs. Perhaps to conclude, I wouldn't say we're quite there yet, but perhaps the direction chosen so far has so far been able to reveal different sorts of kind of lower hanging fruits and things we need to address on a systemic level to be able to achieve the targets and goals set out. And obviously, this builds upon the idea and foundation of crime data, but expands on it in into directions that perhaps make it more actionable and and and something you can utilize in a in a political steering cycle so I guess I'd like to conclude by by saying that safety and security are not simply things that happen to us but but they are something we can first of all prepare for adapt to and also help shape Mr. Chair thank you very much.

Thank you very much, Shari, for giving us a sense of how we can have not only have a better grasp of the future, but also be able to influence the future. That sounds very ambitious, maybe even a little bit optimistic given how much change is happening. But you've given us some tools with which we can do about, with which we can try to shape the future, at least our operating environment. I like your points about forecast, adaption, and shape. I noticed you had used the example of hotspots in Finland. I don't know if that's a reference to saunas, but Anyway, I also like this one slide you had about the foresight turning 2031 impact goals into five year steering questions, which brings to mind to me at least the possibility that since the UN itself, through its Congresses, operates on five year cycles, maybe there would be some way to. incorporate into these decisions not just a discussion of the problems that the UN crime program is confronting today, but perhaps we could have some side discussions on a five-year cycle on, well, what might the situation be five years down the road at our next UN Congress. So, thank you very much, Ari, for this presentation. the, let's see, we'll now go from foresight to the discussion of data management, data security, et cetera. And I have the pleasure to introduce two speakers from the Korean Institute of Crime and Justice, Dr. Chi Tai Huang and Mr. Won Shin Hong. Dr. Shitai Huang holds a doctorate in criminal sociology and is Director General of the Department of Crime and Justice Trends at the KICJ, which is one of the institutes affiliated with the United Nations Program Institute. And he leads the institute's work on the criminal justice statistics database. Mr. Wang Xinhong is head of the Data and Statistics Office at the Korean Institute, and he leads the Institute's digital policy work and turns crime statistics into digital services. So we'll be able to take a look at two sides of dealing with data. So please, the floor is yours.

Good morning, everyone. My name is Jitae Hwang. First name is Jitae, last name is Hwang, from the Korean Institute of Criminology and Justice, KICJ. Our presentation. Our presentation has two parts. I will speak about the crime and criminal justice statistics. Crime and criminal justice statistics is a database system managed by KICJ. My colleague, Mr. Hong Won Shin, will speak about the research knowledge in the age of AI. I majored in criminal sociology and concerned about criminal statistics, and he majored in computer science and concerned about AI. Our title is from criminal justice statistics to AI-ready knowledge. kind of data changes, technology changes, but the principles of data governance stay the same. That is our message today. Let me start with the problem. In Korea, criminal justice statistics are not produced in one place. four institutions produces them. The police record crime occurrences and investigations. The prosecution record case processing and prosecution decisions. The court record trials and judicial decisions. Corrections record imprisonment, probations, and preventions. Each institution produces reliable data for its own mandate. That is not the difficulty. The difficulty is that the same case is counted at different stages, with a different unit each time. Police count an incident. Prosecution count a case. The court counts a dependent. Corrections count a person in custody. So four correct numbers can describe one reality and still refuse to add up. You cannot simply place them side by side. How can data produced by different institutions become consistent, comparable, and usable over time? That is the government's management question we have worked on for 20 for 20 years. This slide shows 20 years of work. In KICJ. In 2006, we began the criminal justice statistics database project. To bring the statistics of different institutions onto one common base. In 2010, we opened up public crime. Statistics portal. Between 2016 and 2020, we expanded the integration to prosecution, court, and corrections statistics. Between 2023 and 2025, we rebuild the data structures and strengthened quality management. And from 2025, we open the data much more widely through national core data and open APIs. In one sentence, we moved from collecting statistics to governing an integrated data infrastructure. Collecting is a task. Governing is a responsibility that does not end. How do we actually do this? It is a managed pipeline with four steps. First, collection from the statistics each institution publishes. Second, validation. We check three things, simple errors, breaks in a time series, and changes in meanings. The third is hardest. So, structuring, it is hard. A published table is designed to be led by a person. We convert it into data that can be carried and analyzed. Fourth, quality management. We maintain the metadata and keep each time series consistent across years. The result is Eight statistical databases and more than 1,500 statistical tables held as one base under one set of rules. The number is not the achievement. The achievement is that a user can now compare across four institutions and across the 20 years without rebuilding the data themselves. Now, standard. When we say standard, we do not only mean a file format. We mean consistent meaning. Six elements must stay consistent: definition, classification, metadata, unit, time, and provenance. Our current scope is national. Consistency within Korean criminal justice statistics. Now I want to be precise on one point, because it concerns an international standard. ICCS, the International Classification of Crime for Statistical Purpose, is not currently applied in our systems, and the reason is structural. not simply a question of effort. ICCS classifies by the act that was committed. Our national statistics classify by legal offenses as defined in Korean criminal law. These two do not line up one to one. A single legal offense can correspond to several ICCS categories. a single ISS category can be spread across several offenses in our law. So our position is this, rather than redesigning national statistics from the beginning, it may be more realistic to build a correspondence table between the two and and to state openly where the match is exact and where is only approximate. Existing Korean statistics, then a correspondence table, then an interpretation compatible with ICCS. That has led to the greater international comparability. Mapping is a possible transition path. It is not something we have implemented, but for any country that already holds decades of national statistics under its own regular categories, it may be the more realistic first step. Consistent data have value only when it is used. Our sequence is collect, standardized, often reuse. First, policy environment. Korean government designated high-value data set as national core data. That designation is not only a promise to publish. It carries requirement on quality, on permit, and on update cycle. So the data arrives in a form that can actually be used, rather than a file that has to be cleaned first. The list already includes weather, transport, real estate, public procurement, health and legislation. Crime and criminal justice statistics are now on that list. Second, how we provide it. We begin inside our own platform, the National Core's data, then open API. And open API means another system can request our data directly and automatically. One boundary here, and I state it deliberately, What we open is aggregated statistics, not individual case record. Openness then protection are not opposition, but the line between them has to be drawn on purpose. Third, the evidence that this works. We open judicial yearbook statistics from 2002 to 2023. structured for time series analysis. Private legal technology company used them to analyze judicial outcomes in case trend as of September 2026 and API received more than 20,000 calls a year. I would ask you to read the number for what it is. These machines calling machines. Nobody is visiting our website. The data is simply working inside somebody else's service. And there, rather than a digital account, is what reuse looks like. So far, structured statistics data. But the AI era raises the same governance question for a very different kind of public data. based research knowledge. I now hand over to my colleague, Mr. Hong Won Shin. Thank you.

Thank you. Good morning. My name is Won Shin Hong. No. Presentation. Sorry. Yeah. And my colleague has shown how KICJ governs and opens criminal justice statistics. I would like to turn to a different kind of data and to the same governance questions. The source is different. The process is different. Here it is semantic structuring rather than validation. The user are different too. they are AI service than the analyst. What does not change is sequence. Establish the meaning, structure it, provide it, and let other reuse it. Different data of various similar governance challenge. A few years ago, we asked how to build an AI service. Then we changed the question and asked instead, what role should a public research institute play in an AI ecosystem? That is a question about comparative advantage, and it gave us a different answer. Building our own service would mean operating a model continuously, improving the user experience, keeping pace with technology that changes every few months, and competing for users against the service that already have them. None of those weaknesses on our side. They are simply not where a research institute has an advantage. What we do have is different. Content that carries public authority. The ability to describe the content's accuracy because we produce it. And an obligation to maintain it for decades, which is written into what we are. A private company cannot easily create those three things. A public institute cannot easily create a consumer AI product. So this is a division of labor, not a retreat. We govern trusted knowledge and we make it reusable by others. So what does AI ready actually mean? It is not converting a PDF into text. Extraction is the easy part. Keeping the meaning intact is the difficult part. We do four things. First, meaningful segmentation. We divide our report by its argument, not by page break. This matters more than it sounds. An AI system retrieves pieces, not whole documents. If a piece is cut in the middle of an argument and the condition is separated from the conclusion, the system will answer confidently with half the meaning. Where you cut the text shapes the answer the user receives. Second, structured metadata. That means title, author, year, topic. and keywords. Third, context. What is studied in what period and where the passage sits inside the product. Fourth, provenance. Every piece stay linked to the authoritative original. Provenance matters most. If an AI uses our research, the user must be able to trace the statement back and check it. Without that link, a correct answer and an invented answer look exactly the same to the reader. The result is ready for machine as well as for people. This slide shows how the data actually moves. Please look at two formats at the bottom. PDF is for people. JSON is structured so that the machine can process it directly, field by field. That change from PDF to JSON barely a little space on this slide. Yeah. Now, what happened in practice, we provide the research structured as AI data. Since launch, our report have recorded more than 7,000 full text view. Sorry. Full text view. I said citations. We would ask you to read the figure carefully because it is not a page view count. There is a different kind of use from someone browsing a website and it changes who read our research. Our reader used to be policy makers and researchers. Now legal practitioners use it as well. They use it inside their daily workflow without being asked to visit a separate public service. I should be careful here. One-punch stewardship is the starting point, not the end. Three steps. Develop. AI-ready research data. Validate. real-world use through a technology platform that told us whether the structure and metadata actually held up in professional use. Then, open the same data through the public data platform. After that, it is available to companies, startups, individual developers, researchers, and students on equal terms. Our principle is short. govern once, reuse many times. A public institution creates more value when where govern data can be reused by many actors than when it sit inside a single service including our own. Let me close by putting the two halves together. Different data, different technology, the same paths. Three lesson for our experience. First, govern the entire data lifecycle, not only collection, but storage, quality, opening, and reuse. Governance steps at collection produce an archive, not usable data. Second, standardize the meaning, not for format. A shared file format does not help if two institutions meaning different things by the same word format makes data readable, meaning makes data comparable. Third, focus public institutions on trusted data and enable broader reuse. We do not need to own every service. We do need to be a source that people can verify and now machines as well. The technology changes, the governance principle remains the same. Trustee data must be on this understandable, connectable and reusable by people, by institutions and increasingly by A. I. Thank you for thank you very much for your attention. Thank you.

Thank you very much, Dr. Ajit. Ty Huang and Mr. Won Chin Hong for this presentation. It's very comprehensive, very major project that the KICJ has undertaken, and I'm sure this will lead to lots of questions and comments. I would like to turn the floor over to my co-moderator, Ms. Leticia Ferhati, who is a member of the Generation Justice Network. Leticia.

Thank you very much for giving me the floor. Thank you very much to Dr. Huang and Dr. Hong for their presentations. If I may draw a few takeaways from this presentation, I think the first is that AI readiness begins with data governance, and the move from criminal justice statistics to AI-ready knowledge is not simply about introducing new technologies, but it requires to address the much more fundamental questions of definitions, classifications, interoperability, and institutional responsibility. Second, criminal justice data is inherently distributed across institutions, so the challenge is not simply collecting more data, but making data generated for different institutional mandates meaningfully comparable. And finally, as data becomes more open and increasingly used by AI systems, questions of access, production, provenance, and accountability become inseparable. Thank you very much for your presentation. I would now want to give the floor to Miss Ottavia Galuzzi, who is an associate expert at UNICRI, where she works on projects aimed at preventing and countering cybercrime and online harms. She brings experience from the public, private and civil society sectors, having worked as a cybersecurity and intelligence analyst and as a counterterrorism and PVE practitioner. Ms. Deluxa, the floor is yours.

Thanks. Thanks a lot, Mr. Chair. Thank you. moderator and thanks a lot for having me representing UNICRI here today. So as mentioned, I'm going to talk about the topic of data privacy and security within the criminal justice sector. And there should be some slides, but I'm also happy to thank you. And the concept of data privacy and security within the criminal justice sector, these are two key concepts strictly interconnected, but oftentimes seem opposite. I'm sure that most of us have heard or dealt with the privacy and security dilemma at least once in our personal or professional life. And this dilemma is the ongoing conflict between protecting people from security threats and respecting their personal rights to keep their information private. which is the right to privacy. In this regard, I believe there are two key areas of friction. First one, and quite known nowadays, is the concept of digital surveillance. So the idea of organizations using tools like facial recognitions or communication tracking to stop crimes, but this can also erode civil liberties and target minority groups, for instance. The second area of friction is in relation to encryption versus law enforcement. So the idea of using strong encryption keeps user data safe from online threat actors, from cyber criminals or terrorist groups, for instance, but it also stops law enforcement from investigating illegal online activities. There have been recent discussion on this point specifically, and the responsibility, for instance, from technology companies to give access to law enforcement investigators in regards to their services and platforms in order to stop and tackle child sexual abuse material. So these are, I believe, two key areas that we see that today we see this dilemma of privacy and security happening. And another area which is not strictly connected to criminal justice sector, but I believe is quite important to mention, is corporate data collection. So there are many technology private companies that would track online behavior to improve user experience or serve their advertisement needs. And once again, here we can see the dilemma between data privacy and data security. But so how can we actually solve this dilemma? We believe that it is important to find an appropriate balance between data privacy and security, which often needs refining, depending on needs and circumstances. In the context of the criminal justice sector, the role of public perception and public trust towards the criminal justice sector, and, often, the focus on law enforcement agencies, is critical. An example that I would like to share here is that, and it might be relatable to many in the room, is that often we don't mind about using biometrics to unlock our phone, but we do mind the use of AI face recognition for CCTV cameras. And while the contexts are very different, the underlying technology is remarkably similar. So what often drives the difference in acceptance is not the technology itself or the data, but who controls the data, how it is used, and whether we trust who is responsible for managing our data. And this really doesn't really lie much on the technology itself, but once again, on the public acceptance. So the public acceptance is not shaped only by a technology can do, but by who is using it for what purpose, under what safeguards and what level of transparency. But so going a bit deeper into what data privacy and data security mean in our world and within the criminal justice sector. So starting with data privacy, this is the principle that Individuals should have control over how their personal information is collected, used, shared, and stored by organizations. There are some key core elements about data privacy, which is consent, control, which means the individuals have the right to access, change, or delete their stored information, transparency, so transparency about who collects the data. It can be private companies or government. and must clearly explain what data they collect and why they need it. Minimisation, which I think is a key component of data privacy, is the fact that organisations should only gather the data required for a specific purpose and nothing beyond it. And something also very important about data privacy, which is not defined as a core element of it, but still very relevant, is the concept of data protection by design and by default. So the idea would be to use technical and organizational measures designed to implement data protection principles. We go back to data minimization. in an effective manner and to integrate the necessary safeguards into the processing in order to protect the rights of data subjects. And all this is to ensure by default that only the personal data necessary for a specific purpose is processed. So here we linked the design to the default. And so we should create technologies that by design at the very inception have data privacy safeguards in place. Moving into data security, which is closer to me as I have a background in cybersecurity per se, data security is often set aside or differentiated from cybersecurity, but actually data security is a key component of cybersecurity. And while cybersecurity normally is described as the collection of tools, policies, security concepts, safeguards, guidelines, best practices, and I could continue, that can be used to protect the organization and user assets, I think that a stronger description of cybersecurity should be that cybersecurity is a mindset. So it should be a personal and organizational mindset to protect yourself and the work you do. every day, particularly when it comes to the criminal justice sector. But so to understand data security, we oftentimes looks at the CIA triad, which is a basic component of cybersecurity once again, that has three main components. So confidentiality, which is the efforts to make sure that data is kept secret or private. Integrity, the efforts to make sure data is trustworthy and free from tampering. and availability, which means that the system, networks, and applications must be functioning as they should and when they should. And this means the same for data, because the data is collected and processed within the system, networks, and applications. And these concepts from data privacy and data security are the foundation to ensure that data remains a reliable, ethical, and actionable source for crime prevention and criminal justice. But being aware of the privacy and security dilemma doesn't mean that we should pursue a one-size-fits-all approach. But as mentioned at the beginning, it's really key to always reassess the dilemma and the interests and needs at stake. And so taking the example of before, of CCTV and facial recognition, how can the security need of using this technology to help law enforcement agencies identify suspects and locating missing persons ensure the privacy of the rest of the individuals that their data will be collected in these CCTV cameras? This is only possible if privacy safeguards are in place, and actually they must be. So we're talking about limited and specific and lawful purposes. The data retention periods must be clearly defined, and the data that is not linked to an investigation has to be deleted or anonymized after a specific amount of time. So as mentioned before, what I think is key in this context is also how criminal justice actors communicate to the public about their use, in this example, of facial recognition, and ultimately pursue a transparency-first mindset. rooted in clear communication, effective public engagement for a sustainable public trust. This, we believe in the work we've been doing at UNICRI, that's the only way to, in a way, solve this data privacy and security dilemma. And so next, I would like to share a few insights into how at Uniqry, we seek to strike, we try to strike an appropriate balance between different and sometimes competing interests related to data privacy and data security. A quick recap on how we work at Uniqry for those who don't know us. So we conduct action-oriented research, and this is grounded in data and statistics. And based on this, we seek to lead evidence-based programming that could be capacity building or to policy to support. And our work, as many in the room, relies on a wide variety of data related to crime prevention. And this can be criminal patterns and trends, including victims and perpetrators' data. Our approach is to address specialised niches and selected topics. As you can imagine, this requires us to always reassess the way we collect and gather data, while leveraging best practices of data privacy and security. For instance, the way we collect and use data for a research project on environmental crime might not be the same as for a project on cyber-enabled fraud. The data points will be different. and the beneficiaries involved most likely will be different. So it's important that we consider different considerations when we look at the data privacy and data security related to a specific project. And so starting from a general example on how UNICRI connects research to practice, so meaning how we go from data-driven findings to identifying areas for focused research and programming. I would like to share a few words about our latest crime and justice study, which is a global interregional study of crime and justice trends, which heavily relate on data once again. And if of interest, this will be presented in a side event tomorrow. But so here, the data-driven findings from this global analysis provided us the basis to identify areas for further research and hopefully programmatic activities in the future. And one of the key areas identified was emerging scam center style ecosystems in Africa. So this is currently an ongoing research. And what I think, and I just wanted to bring this example to you all, because it's a good practice on how we try to strike an appropriate balance between different and sometimes competing interests. So in this research, we have, on one hand, survivors holding the most valuable data on operations, locations, and methods, while also being most at risk from exposure. And on the other, the safeguards we design around that. And namely, that would be consent under UN personnel data protection and privacy principles of 2018. But also, we approach this research with a need-to-know basis, meaning that the data storage system limits access to internal and external staff on a need-to-know basis. For instance, there's a few of us working on this research, and not even all of us have access to survivors' and victims' data, because perhaps there is not the need to do so. And finally, anonymization is a big component of this type of research. along with consent-based recording, but also identifying details such as routes and sites related to a scum centre. They are aggregated before publication. The message through this specific research, but I would say all the research that we do, is that data itself is unproblematic. The key is to ensure that data remains a reliable, ethical and actionable resource for crime prevention and criminal justice. And this becomes even more important when we talk about AI, as different colleagues mentioned already this morning, that we should really learn to process only the data that is strictly necessary within AI. And in this regard, there is ongoing conversation in promoting the use of a frugal approach with regards to the AI, so only to leverage the necessary data, which is oftentimes called smart data. versus big data, so large amount of data, which is often unnecessary. And I think this will be on my end. Thanks again for your attention and back to you, moderator. Thank you.

Thank you very much, Milica Huczi, for an excellent presentation. I really liked your point about crime being increasingly interconnected across borders and markets, technologies and its rapid adaptation and how central challenge for criminal justice is to ensure that institutions, cooperation mechanisms and legal frameworks can adapt at a comparable speed. I now would like to give the floor to Matif to continue the moderation.

Thank you, Letitia. We have started looking at these issues of data management, data security, and data privacy from the perspective of one country, Korea, that we have worked up to uniquely. I now have the pleasure to give the floor to Salome Flores Sierra from the United Nations Office on Drugs and Crime to provide us with a more global view of the work that the UNODC has been doing helping member states around the world in this. She has been working with the program office of the UNODC in Honduras. Salome, we're very much looking forward to your presentation.

Thank you very much, dear moderator, chair, good morning, distinguished participants. My presentation today is about the global standards that UNODC has developed together with many member states. And, well, I think it fits very well to talk about global standards after yesterday's session and after the interventions today, because we see how across the regions there are different levels of maturity when it comes to data. In some countries, data collection is very basic. In other countries, as we saw, there are statistical projects in place, some efforts to standardize administrative data from different criminal justice institutions. In other countries, there are national surveys to understand victimization, to understand violence against women, corruption, different issues. Those are the traditional data sources to inform crime prevention policies. But we also heard about the need to use alternative information, non-traditional data sources, sources from the private sector, information from social media, from internet. So we have a very diverse landscape when it comes to data. And of course, there's no single data set that can inform policies. Usually, as we saw from the presentations yesterday and this morning, there is data all over the place. There is a fragmentation when it comes to data. So how do we start? Yesterday, someone was saying in the presentations that we need a basic common standard to start. And this is what my presentation is about. It's for looking at some of the tools that, like I said, UNODC has been developing together with many countries, and that we are trying to support countries to adopt them. So little by little, the information that is available from the criminal justice system is not only shared, but also disseminated in a transparent way. But more importantly, that is actionable data for policymakers to really understand crime trends and new threats. So very briefly, Why global standards? Or how does the world looks without global standards? And we heard already a lot about this. Without global standards, it's difficult to structure data from different criminal justice institutions. So it's difficult to integrate data and to compare data even within one country. We see how different criminal institutions use different recording systems. They can also use different definitions, different categories and variables. So it is difficult to integrate information. So without global standards, we cannot really sort out this initial challenge. And of course, if we don't have this, we cannot detect emerging threats. And then finally, without global standards, we cannot monitor global commitments like the Sustainable Development Goals. And as you know, there are a number of indicators, especially under SDG 16. that are connected to peace, security, and strong institutions, and that are usually coming from administrative data or from surveys that are the traditional sources to inform crime prevention. But they are still there, and some countries are reporting their progress on improving SDG 16 targets. And that's why global standards are useful. So with that, I will move to, well, maybe I'll just highlight the importance of global standards to improve the consistency across different data sources. We have structured data, but also with these new alternative data sources, we have a lot of unstructured data. Global standards are useful for both, for making sense of the information out there, to improve, like I said, consistency, comparability, and overall quality of data. And of course, if we have good actionable data, that would be for the benefit of policymakers, but also for the public to understand what is in relation to crime and new forms of crime. So in this slide, you can see some of the statistical standards and some of the methodological tools, like I said, that have been developed by UNODC. You see at the top there, the International Classification of Crimes for Statistical Purposes, ICCS. This has been mentioned in several presentations as the backbone for integrating criminal justice statistics. This is a classification that was endorsed by the CCPCJ and by the UN Statistical Commission, and it's accepted as the classification to start making sense about data on victims, about data on perpetrators, and the context and which crimes occur. And following the ICCS, we have other tools like the ICTIP, that is the International Classification for Administrative Data on Trafficking in Persons. That is a new standard that hopefully will be endorsed by the UN Statistical Commission next year. And then we have another framework, the statistical framework for measuring femicide, that is useful for registering gender-related killings of women and girls, especially in those countries in which femicide has not been criminalized yet. Finally, we have another framework to measure corruption, which is, as you know, very complex phenomena that should be tried to measure by collecting different information and not just one or a few indicators. And then we also have a conceptual framework for countries to measure, collect data and measure illicit financial flows. And those are the statistical global standards that we have been developing with other countries and, like I said, supporting them to use them. And then we have another set of methodological tools that are more connected to surveys, well, and administrative records as well. So we have a handbook on governance and statistics that was jointly developed with UNDP. a manual on corruption surveys, a manual on victimization surveys, because as we saw, administrative data alone is not enough to inform crime prevention. And we do need to bring in the perspective of the victims with information that is coming directly from them about the experiences that they have, about the perception of safety, the perception of effectiveness about the criminal justice institutions, and of course about the reporting behavior, if they report their crime and if in case they don't report it, the reasons why they do not report the crimes. And then a couple of guidelines there, well, some guidelines that I will talk about in a second. And then finally, a survey initiative to measure different indicators on the Sustainable Development Goal 16. So Just to go into more detail about the ICCS for those that do not know this standard, the ICCS basically establishes a common language for crime statistics and is useful to overcome the challenge of using different penal codes. As we know, every country has its own penal code. And how do we make sense of data coming from all these different countries if they all have typically, if they all use typically their penal code to register, to record crimes. And based on several consultations across all the world, we came up with this classification. It's a way for countries to define and record crimes based on common definitions. Basically, these common definitions are based on the actual behaviors that constitute an offense. This classification, like I said before, was endorsed in 2015, is the first iteration. And for those of you that are not familiarized with classifications, classifications evolve over time. Just to give you an example, the classification for diseases has 11 iterations, and if I remember correctly, was first created 100 years ago. And this is just to say that just like with diseases, crime is also kind of a disease and it evolves. And we have a classification now, but as we know, there are new crimes that are appearing and we... will continue working ideally to update this classification to fit the new crime landscape. So the use of classification, like we saw yesterday in the presentation by Morocco and this morning in the presentation by the Korean Institute of Criminology and Justice, this classification is useful to start integrating data in a way that makes sense. And I have this example here. This shows how the classification looks for homicide data. The classifications has different elements, sections, and categories. And in here, we see, for example, that we have section one that corresponds to acts leading to death or intending to cause death. And we have here the definition of intentional homicide as per the international crime classification for statistical purposes. And one key element of this classification is that we have a list there on the right hand side about inclusions and exclusions. And this is really what makes comparability possible across jurisdictions because of as we know there are different classifications, there are different definitions at the national level but when recording or and more importantly when let's say reporting to UNODC through the UN crime trend survey, countries are using this way of classifying to make sure that we have a comparability that we are not comparing apples with pears but we are basically trying to make sense and the fine apples. It sounds difficult, we know, because of course there's no expectation of countries changing their penal codes, but using this kind of classifications is useful for statistical purposes. And this is how the ICCS translate national offense definitions into globally comparable crime statistics. This is just an example of the homicide rate by region and globally. You see the different rates there across the region. This comes from a recent report that was launched, the SDG 16 Global Progress Report, Peace, Justice and Inclusion. It's available there. But just to give you an example about how to make sense of the data across the regions, but also, like I said, nationally. And just before I go deeper into the other frameworks, just to say that the ICCS, I mentioned the definitions, but one important aspect of the ICCS is that it also includes guidance about information or categories and variables for victims, for perpetrators and for characterizing the context in which crimes occur, which is information extremely valuable to inform policies. Now, I was mentioning the statistical framework for measuring the gender-related killings of women and girls, femicide. the you can access the publication in the QR code that is there in the slide. But basically through this framework, we can facilitate or enable countries to collect data on on this phenomena, even if it's not available. I mean, if it even if has not been criminalized nationally. And this is because by using homicide data and by using specific elements, countries can record and understand the intentionality of the killing and understanding if this can be considered as femicide. In this slide, we have examples of the application of the statistical framework to measure femicide. These are examples from France. And in here, you can see on the left-hand side the different types of motivations of the gender-related killing. And on the right-hand side, on the graph there, we can see who was the person that committed the crime, which is also extremely important to understand the conditions or the situation in which these crimes are happening, but also who are the common perpetrators of this really horrible crime. Then, moving forward to the other global standards, we have another conceptual framework for helping countries in measuring illicit financial flows. This framework was developed jointly with UNCTAD, and this is also useful for countries to report on the SDG connected to measuring illicit financial flows. The framework has this distinction between the illicit financial flows associated to illegal activities. This is illegal markets like drug trafficking, like trafficking in persons, like crimes in environment. but also those illicit financial flows associated to legal activities. These are more connected to illicit tax and commercial practices like tax avoidance. And yeah, you can also access the publication there. It's available in different languages. Some countries have been piloting this framework, but as you can imagine, measuring illicit financial flows is still quite challenging, but the framework is available there for you to use. Then we move to another set of tools this time to measure corruption. In this slide, you see the four components there. If you visit the website, you will find different publications. We have the manual on corruption surveys to understand directly from victims if they have experienced corruption and in which sectors and what was the modus operandi of this crime. But like I said before, we also have new tools, a statistical framework that offers a menu of indicators on corruption to explore or to measure different types of corruption, to also understand risks associated to corruption, to monitor and measure the effect of preventive measures, and finally, to also understand the environment to report and address corruption. And we also have a practical guidance to implement this statistical framework to measure corruption. And finally, almost the end, we have a set of guidelines that are a common framework for producing criminal justice statistics. As you can see, there are four publications, four set of guidelines. They are, each guideline are different, or some are dedicated to police, others for prosecution and courts. And finally, we have the publication on PRESENCE. And another one that is the one that you see on the slide there at the end, that are the guidelines for the governance of a statistical data and criminal justice system. So this is basically for countries to have their own national arrangements around criminal justice data. If you look at the guidelines, you will find different recommendations, again, about definitions, categories, variables, and even indicators that are relevant for each one of these authorities and to make sense also of all this data. And more importantly, like I said at the beginning, to use the data in a way that is actionable for policymakers to understand what is happening and also to use it for decision making. Finally, this is the last one, the survey guidelines. We heard yesterday from our colleagues in Morocco that they've been using the manual on victimization surveys. This is kind of an old manual that was jointly developed between UNODC and the UNECE, the Economic Commission for Europe. and still being very widely used for countries still conducting victimization surveys as a key complementary source to inform crime prevention policies. We also have the corruption survey manual there, and finally the SDG 16 survey initiative. And all these Manuals or survey guidelines also are connected to the international crime classification for statistical purposes as the backbone of all crime statistics. And with this, I finish my presentation and really looking forward to the discussion. Over to you, moderator.

Thank you very much for your excellent presentation and specifically for shedding light on the importance of the interplay between statistical data and how much this is used fundamentally in crime prevention, criminal justice, and most importantly, how data needs to be actionable to achieve those outcomes. I think I can now hand over to Matti, who will moderate the next part of the session.

Okay, thank you, Laetitia. We'll go now into the interaction, and I would like to begin this interaction by asking if any members of, or any people here up on the podium would like to ask questions or comments on the presentations given by the others. You're all experts in this area, so please feel free to weigh in with questions or comments. Okay. If there are no immediate questions or comments, I'd like to start by asking Ari a question, but at the same time, I would like to ask our Korean colleagues a question. I was fascinated by the amount of work that has gone into what you are doing in Korea. in gathering data and in managing data, et cetera. And I have a political question. How difficult was it to launch this project? Because it requires a long period of time. It requires a large investment of money. Was there something that sparked this decision to develop this comprehensive national system? or was it perhaps driven by the Korean Institute, or where did the initiative come from? How difficult was it to make the political move, because it does require a political decision to do such a massive thing? And a follow-up question to that, if you don't mind, is Korea is a fairly wealthy nation. How difficult would this be to replicate in another a nation that perhaps is not as well developed in research infrastructure. So but while you think about this question, Ari, if I could go over to you, you had given some examples of the use of foresight, which I find fascinating. What you're trying to do is look, you are saying basically, as you said in your presentation, that crime data is a rearview mirror that will tell you what has happened before. And you gave some examples from Finland about the development or the use of foresight. Would you have any examples from other countries of either small or large use of foresight, a kind of aha moment? This is something that we should be paying attention to. Harry, would you be so bold as to try to respond?

Yes, of course. My pleasure. What comes immediately to mind, of course, is the strategic foresight in governance and government that has been done, especially in Singapore, where it's a very systemic approach, kind of, how should I say, cross-silo, cross-administrative approach, and there's even strong signs of it being of a leading North Star to all of policy. So that definitely has inspired us. We partnered during our process, we partnered with, and regardless of its name, it's an international consultancy, government consultancy called Demos Helsinki, who who were kind enough to bring in international experience and expertise. On a similar note also, or perhaps kind of moving forward with that, interestingly, using foresight and scenarios isn't in within crime prevention and law enforcement, isn't really kind of that far away. Namely, the intelligence community has been doing this for quite a while. And obviously, as we know, intelligence is quite close to crime prevention and criminal policy. So we oftentimes perhaps even go farther, farther, far from home, so to speak, when looking for answers. And that's obviously the approach within the intelligence community is something we also domestically adopted and utilized. But again, perhaps an added layer to that definitely was looking at the positives because also within the intelligence community, it is quite threat based. And I can't enough emphasise perhaps the need for looking at positives as well, going back to my point on our strategic reports being read politically as scare tactics for more funding. So we simply needed to do it for credibility's sake, because if everything is always a threat and we always need money, as you very well know, it doesn't really kind of pan out in the political progress. So we have to be able to communicate, well, these things are being replaced by these these other things. I perhaps took the liberty of kind of expanding within my answer. Sorry.

Thank you very much, Ari. You've taken us over to the other part of the world, which is always useful to get a different perspective. May I ask if our colleagues from the Korean Institute would have a response? And I understand this will be interpreted.

Thank you. Two short answers. It started as Research need, not a political decision. The hardest part was not budget. It was trust between institutions. What helped is that KICJ is a research institute. We do not supervise the agencies, so nobody felt inspected. And yes, I believe other countries can do this. Expensive part was not technology. It was people who understand both statistics and criminal justice. Started with one indicator and one time series. Make it consistent. Show it is useful. Then add the next one. small system that people trust is worth more than a large system that nobody uses.

Thank you.

Thank you very much. And I would now like to turn the floor over to our chair to see if we can stimulate some response from the audience.

Yeah, thank you. I think this was really a lot of food for thought, a feast for thought, so to speak. And we have some time for interaction. And yeah, who could I give the floor? I see the United States. You have the floor.

Thank you, Chair, and thank you to the panelists. Really great presentations. I have a series of comments to make and a question at the end, both in response to the panel this morning, but building upon some of our reflections from yesterday. Part of what data should do is tell us what the most urgent threats are. Data should be driving our priorities and our interventions. If everything is important, then nothing is important. Unfortunately, we have a declaration that is far too lengthy, drafted by Vienna-based generalists with no real direction, no prioritization. The fact that drugs gets one paragraph and alien smuggling gets one paragraph is alarming. How much more progress can we make if the Congress had a one-page list of priorities backed by data that focuses the world on our most urgent threats? Furthermore, our actions should be grounded in our treaty-based mandates. Of all the side events this week, not one of them is looking at the implementation review mechanism to the UN Convention against Transnational Organized Crime. If you and ODC's slide on methodological tools, the IRM was also absent from that slide. The IRM is incredibly slow, stagnant even, and we're incredibly behind. which we have to attest to political will. States are not funding the IRM and they're not demonstrating the will required to demonstrate their implementation of the convention, a convention that has been around for decades. The IRM should be a valuable tool and means by which states can include data and evidence of progress. That said, we submit some recommendations. Number one. The next Congress should have a more refined and streamlined set of priorities backed by real data to drive outcome-based international action. Two, the Congress should urge greater progress in the review mechanism and urge data-backed evidence be incorporated into the review mechanism. I made an intervention yesterday that this Congress is void of any data to look at or analyze. States did not come to the table with any data. So to clarify, we also submit the following recommendation for this workshop. States should submit to the 2031 Congress concrete, quantitative, and qualitative data identifying the most urgent threats, as well as demonstrating the results of national efforts preventing and disrupting crime. In response to some of the substantive issues brought up by the panelists, the Trump administration is opposed to efforts by foreign governments or the UN to weaken commercial encryption, bypass technical architecture protecting U.S. citizen privacy, or otherwise exert pressure on the design decisions of U.S. technology companies. The United States is committed to holding an open, interoperable, reliable, and secure cyberspace. Encryption anchors the international community's trust in this system and enables trustworthy cross-border data flows around the world every day. Our modern digital economy is predicted on the existence of strong encryption. The United States underscores also that it does not support the Sustainable Development Goals and objects to their inclusion in the recommendations from this workshop. Data governance is best managed and decided by states themselves in line with national sovereignty. Regarding so-called illicit financial flows, the correct terminology per treaty language is illicit proceeds of crime, and the report from this workshop should reflect the treaty language. Lastly, a question to the panel. Thanks for entertaining the long intervention. How can we leverage the IRM as a tool for data collection and analysis? Thank you.

Thank you. And of course, it's not part of my role to make substantive statements here, but as I was part of the process that led to the Abu Dhabi Declaration, I'd like to say that I think for the vast majority of delegations, the declaration is a big success given the challenging political environment. And now I'd like to turn to panelists if they would like to react to the questions raised. I see no big appetite, but the moderator.

Yes, thank you, Mr. Chairman. I'd also like to thank the distinguished representative of the United States for his statement of position, but also his thoughtful suggestions and recommendations. I would just like to jump in on one point, and that is the implementation review mechanism. You were speaking specifically about the implementation review mechanism for UNTOC, if I understand correctly, which is a fairly new one. And I do remember the lengthy, if I can really draw out that lengthy discussions we had on what this should look like. And there was a need to balance the availability of funding with what could be done, which always comes up. And it was also pointed out that compared to UNCAC, the United Nations Convention Against Corruption, UNTOC is much vaster. So there was also difficulties in trying to assess what would the priorities be there. Another point that occurs to me is I appreciate your interest in getting more attention to the implementation review mechanism and in getting more data into that. I'm a little bit hesitant as to whether the UN Congress would be the right place, because I do sense that there has been discussion in Vienna between what is the mandate of the UN Congresses themselves, which basically represent the entire global society, as opposed to the conferences of the state parties. But I fully agree with you, and this is simply my position as a member of the delegation of Finland, that we do need more data. We do need more data that can provide a more solid base for making decisions within the context of not only UN congresses, but also UN commissions, et cetera. I would then add on a personal note that I have been looking very, very closely at the information produced by the implementation review mechanism for the United Nations Convention on Corruption. And I have found that extremely fascinating, not because there was so much statistical data provided, although there were in almost all of the reviews of the individual member states that are parties to UNCAC, but in the qualitative data provided in those various reports. And I have to say that I was daunted at first at looking at the amount of information produced by that mechanism But I do believe that the UNODC has been able to harness that quite well. And if there are very specific issues that are of interest to people regarding corruption, there's a lot of wealth to be found there. I apologize, Mr. Chairman, for misusing my moderator's role.

No, no, no. Thank you very much. And I think also Salome Flores would like to add some remarks.

Thank you, Mr. Chair. Just to say that UNODC is mandated to collect data from all member states through different or specific dedicated data collection tools. We have the UN Crime Trend Survey that is sent out to all member states every year to collect data on crimes and the functioning of the criminal justice system. We have another dedicated set of tools for collecting data on drugs, seizures, and the different aspects of drugs, demand and supply. We also have the dedicated questionnaire to collect data on trafficking in persons. We have another one to collect data on firearms. So the tools are there. UNODC launches these requests of data to all member states every year, but unfortunately the reporting or the data sharing by member states is not always consistent. And actually yesterday there was also an exchange about how difficult it is to measure trends if we have gaps in the existing data. So maybe this kind of workshop is a relevant opportunity to actually remind ourselves of these existing tools to collect data globally across different crime issues and invite all member states to share their data so it's available for everybody. UNODC has a data portal and the data that is collected every year is available in that data portal but again there are gaps and it would be very useful to get data from all the member states really so we can actually inform not only the next Congress, but all the governing bodies about what is happening with crime. So, indeed, also to say that the UN does not decide on its own what to do, but it's actually whatever the UN is doing responds to what member states agree upon. So, indeed, enriching all the implementation review mechanism of the UNTOC, but also other implementation review mechanism and mechanisms and other tools for monitoring progress would be extremely relevant if whatever findings we have are enriched with data on a regular basis. UNODC would be very happy to support countries in improving existing data and also would be very happy to continue working in making sense of all the data that is received. Because, like we've been discussing, there's a lot of data out there, structure, unstructured, sometimes it's a matter of strengthening existing efforts at the national level, improve coordination across different criminal justice institutions, improve existing arrangements for recording data, sharing data, enabling interoperability, improving capacities to analyze existing data, make sure that the data is available there, that dissemination is done in a way that is easy for everybody to access and also to understand. So with this, over to you, Mr. Chair.

Thank you. Are there any further requests for the floor, questions, comments on the first session of today's meeting? I see none. Then we proceed to the second part to the second session of today's meeting and hand the floor back to Mr. Jorssen.

Thank you, Mr. Chairman. We'll now shift gears somewhat. We've looked at foresight, thanks to Ari. We have then looked at data management, data security, and basically the data itself. Now we will into the second-half where we are going to examine various new aspects of international cooperation in data and data sharing for better national and international policy. Our first speaker is Ms. Israa Mohamed Abdelmagd Khali Mohamed, who is a member of the Generation Justice Youth Network. She holds a bachelor's degree in economics from the Faculty of Economics and Political Science at the Alexandria University in Egypt. She's worked previously, sorry, currently working as a researcher at the Egyptian Money Laundering and Terrorist Financing Combating Unit. which is basically the Egyptian FIU, the Financial Intelligence Unit. She also represents the Egyptian FIU in the Terrorist Financing Working Group of the Counter-Terrorism Information Exchange and Criminal Justice Responses project implemented by the European Union Agency for Law Enforcement Training. So she will be able to provide us with a good view on international cooperation in an extremely important area. Isra, the floor is yours.

Thank you, moderator, chairman, distinguished participants. At the outset, let me briefly introduce myself. I'm Israa Abulmagd from Egypt. working at the Egyptian Money Laundering and Terrorist Financing Combating Unit. So I start my presentation with new recent findings by international organizations, such as the Financial Action Task Force, that show how the threat of organized and financial crime is moving faster forward and further and increasingly through digital channels and technologies. I started with fraud. Today, fraud is not a marginal financial crime issue. FATF reports that 156 jurisdictions, or 90% of these jurisdictions it assessed throughout the fourth round of mutual evaluation, identify fraud as a major money laundering risk. And also, just weeks ago, FATF published new findings on underground hawala and other similar service providers. It indicated that more than 80% of responding jurisdictions to FATF report identified these systems among the principal channels used for professional money laundering, while nearly 70% of reported countries report the integration of new technologies, which is increasingly referred to as digital hawala. I have to highlight another emerging area, which is the online gaming and gambling. FATF's latest report published this month, based on responses and inputs from 80 jurisdictions, found that these activities are becoming increasingly online, cross-border and interconnected, with multiple payment methods such as e-wallets, virtual assets, which make it difficult to trace the illicit financial flows behind it. We also see new challenges in terrorist financing investigations as FATF found that 69% of jurisdictions it assessed through the fourth round of mutual evaluation showed major structural and deficiencies in effectively investigating, prosecuting, and convicting terrorism financing cases. Also, such deficiencies have also recently been highlighted by UNCTAD. Also regarding crowdfunding, crowdfunding should be for good and charitable purposes. However, it could be misused for illegitimate purposes, specifically for terrorist financing. Also, FATF cited more than 6 million crowdfunding campaigns worldwide only in 2022, which highlight the volume and variety of this activity that can make illicit activity even harder to be detected. According to FATF report also published in June 2026 on the use of social media, instant messaging application and streaming platforms, especially for terrorist financing, less than 30% of jurisdictions that contributed to the report indicated they cover terrorist financing risk related to these platforms in their national risk assessment, and 38% considered it to be moderate risk. Jurisdictions facing conflict or active terrorism threats generally reported the risk to be high or very high. So now I have to highlight the illicit financial flows. As criminal activities become increasingly diverse, sophisticated, and transnational, The methods used to generate and move illicit proceeds are also becoming more complex. These developments contribute to illicit financial flows, which can move rapidly across borders and through different channels. This creates several challenges for measuring illicit financial flows, and among them that there have been evolving concepts and definitions across international frameworks and the only unified definition for IFFs was the definition developed by the United Nations. and it was for measurements purposes. Also, second, measuring these flows remains challenging because they are often concealed while the relevant data may be fragmented, incomplete or unavailable. Also, the information needed to identify and understand IFF distributed across different authorities and jurisdictions. Consequently, countries need to recover their assets. And FATF's fourth round of mutual evaluation showed that More than 80% of jurisdictions are operating at low or moderate levels of effectiveness in asset recovery. This is, by the way, the effect, the immediate outcome 8 from the standards of FATF. Also, the international community keeps trying to address these challenges. As the UNODC representative have highlighted that IFFs are incorporated into the sustainable development goals through indicators 16.4.1, which measures their total inward and outward value. UNCTAD and UNODC have also developed a statistical framework to support their measurement. In addition, complementary initiatives, such as the Global Organized Crime Index, can help us understand the criminal networks and activities that may generate illicit flows. The key point, therefore, is that measuring illicit financial flows is not simply about following the money. We need to follow the money. We need to do parallel financial investigation. But also, it requires understanding the criminal activities behind those flows and connecting relevant data across authorities and across borders in order to identify the highest risks and the overall risk, national risk, that I will explain furthermore in the next slides. eventually to put a frame preventive measures to medicate the identified risk and threats. Let me now turn to the national risk assessment that comes in the line with countries need to identify, assess and understand the money laundering, terrorist and proliferation financing risk within their jurisdictions and then take action and apply resources to mitigate such risks based on the risk-based approach which comes in accordance with the FATF standards specifically recommendation one and immediate outcome one. As you may see in this figure, this slide, risk is a function of threats and vulnerabilities. So what is the purpose of NRE? In simple terms, it's about bringing information together to understand where the risks are and what needs to be done about them. First, we look at the threats. In general, it's a person, group or activity with potential to cause harm to state, society, or the economy. Then we look at the vulnerabilities where criminals may be able to exploit weaknesses in different sectors, products, or services, or channels to facilitate their illicit activities. By looking at the threats and vulnerabilities together, we can assess the level of national and sectoral risk. identify the highest risks area. Importantly, NRA or National Risk Assessment shouldn't be seen as a one-time exercise. It should help countries keep their understanding of risk up to date and identify new and emerging forms of crime as criminal methods and technologies continue to evolve. Finally, the value of NRA is not simply in collecting more data, randomly data, but in connecting and analyzing data to build a clearer and more up-to-date understanding of risk, identify emerging threats, and support timely and targeted actions. So, NRA also requires cooperation among competent authorities. All authorities should to cooperate with each other. Law enforcement, FIU, judiciary and private sector, including financial institutions and designated non-financial business and professions. Also, NRA requires the integration of information from multiple sources. For example, as such as transaction reports, investigative and criminal data, also open source or OSINT. The online published data, including international reports and assessment, we have to get the ultimate benefit out of it. And definitely the questionnaires and surveys made exclusively for competent authorities. And it's very important to know each sector's threats and vulnerabilities. From the forementioned slides, we can conclude that crimes in nature are cross-border and effectively combating them needs reliance on multiple sources of data, both national and international stakeholders. Hence, There is an urge for international cooperation from formal and informal channels because the cross-border nature of illicit financial flows makes it essential to trace assets and transactions, obtain information and evidence in order to freeze, seize and confiscate criminal proceeds. And now to the mutual legal assistance. This is one of the formal mechanisms for international cooperation. It provides a legal framework for obtaining evidence and assistance, specifically when coercive measures or judicial evidence are required. However, there are some challenges. MLA can also involve some challenges, such as lengthy procedures. For example, there is competent authority or through diplomatic channels, the different legal requirements and the language barriers make it difficult and cause delays where requests are complex or incomplete. Consequently, this highlights the importance of informal international cooperation and practitioner-based cooperation, which can provide faster and more flexible channels alongside with MLA. Cooperation among competent authorities, including FIU to FIU cooperation through bilateral memorandum of understanding and the agreement group. It facilitates the secure exchange of financial intelligence and support cross-border investigations. Other networks, such as the IRNs, for example, we are in Egypt, member of the MENA IRN. It support informal cooperation on asset tracing and recovery. The GLOBE, I have to highlight this, the GLOBE network established with the support of UNODC, strengthens cooperation among anti-corruption law enforcement authorities in cross-border cooperation cases and related offenses, including investigation, prosecution, and asset recovery. The key point, therefore, is to use formal and informal cooperation together. They are not competing channels, they complement each other. informal channels to develop leads and exchange information quickly, and also MLA, where formal evidence or judicial measures are required. I'll show a case study. And to move from the theory to practical case, I'd like to present a case that elaborate more to answer one question, which is, what can data tell us if you connect the dots? So Actually, because to elaborate further, it was that two suspects acted in, arrested in act. Police found some items with them. And then at this stage, we have the items were weapons and some cash. Also drugs. At this stage, we have drug case, but we already know the full-- we don't have the full picture. Further intelligence by the General Directorate for Drug Control and Ministry of Interior showed that the two suspects together with others were part of criminal group, including a foreign national. Some members also traveled to foreign countries. So that point, we had to make the parallel financial investigation. The general directorate for drug control requested from us the financial intelligence unit to conduct parallel financial investigation about the suspects and other of the criminal group. To conduct our intelligence, we requested information from private sector, mainly banks operating in Egypt regarding those, and we consulted real estate. and also we requested information through admin group. And this is where the picture started to change. Some of the suspects bank accounts showed cash deposits followed with by withdrawals and also the international cooperation showed that. The other counter part reported that one of the suspects was a beneficial owner of a business in foreign country. The suspects also acquired two cars and residential units. This is very important point. Keeping up to date beneficial ownership information can help authorities identify person actually behind the activity rather than the proxies. Based on the available evidence indicating the commission of a predicate offense, the competent authorities requested provisional measures. This is the public prosecution ordered the seizure of the suspect's assets. So what does this case tell us? First, illicit funds do not necessarily enter or remain in the formal financial system. So parallel financial investigation is very important. That can reveal the financial flows behind the criminal activities. Also, this requires strong national coordination and inter-agency in order to connect information held by different authorities. And finally, when the assets or criminal actors cross borders, international cooperation, either formal or informal, becomes essential. to access reliable and up-to-date beneficial ownership information. So the message is simple. Data becomes powerful when we connect it across agencies, systems, and across borders, and eventually analyze it to identify the criminal patterns and typologies. And that's from me. Thank you so much for your attention to you, to the moderator. Thank you.

Thank you, Isra, for your presentation. You have provided a very good overview of the issue. You have added some examples, basically the issue of international financial flows, stressing the importance of national risk assessment, identifying threats, vulnerabilities. and also the need for risk mitigation efforts. But you've also brought us up to the international cooperation level, adding also the importance of not only formal cooperation, but of informal cooperation, which is often so, so much quicker and more efficient. Well, thank you. Our next speaker represents INTERPOL. Mr. William Hippert, who is the Director pro tempore for Operational Support and Analysis at INTERPOL, where he oversees international criminal analysis and operational support to law enforcement bodies worldwide. In his former life, he was a French police commissioner and he has more than 17 years of experience in law enforcement and has held senior positions in both operational and headquarters environments, specializing in organized crime, transnational drug trafficking, financial crime, intelligence, and international police cooperation. We're very much looking forward to your insights on the trials and tribulations, but also the pleasures of international cooperation. William, the floor is yours.

Thank you very much, Mr. Chair, moderators, distinguished delegates, colleagues, ladies and gentlemen, good afternoon. That's the four official languages of Interpol. As I will likely be the last panelist of this session, I hope you still have a bit of energy for me, especially before lunchtime. I don't have any slides. 168 million That's the number I would like to start this session with, 168 million. That's the amount of data available in Interpol databases as we speak right now from our 196 member countries. And that makes us, that makes Interpol, since its very beginning in 1923, the world's largest police data bank. But our role is not limited to that, to my opinion. Interpol is as well, and maybe mostly, a hub connecting critical police information, police data, from our member countries. And this amount of data has been searched just to provide you with other numbers, and then I would stop with figures. All of these numbers, all of this data has been searched last year during 2025, 9.2 billion searches last year, billion times. Names, travel documents, biometrics data, child sexual abuse materials, phone numbers, IP addresses, stolen vehicles, nominals, weapons, bank accounts. all related to organized crime and terrorism. That is around 300 searches every second. And I would like to ask one question. Why do nearly 200 countries share sensitive information with us, or not only with us, but between them? After nearly 17 years supervising investigative teams in my country, France, and I'm still French police commissioner, I know one thing, international cooperation is not an option. You have to choose the right channel, understand how your partner works, thinks, what is their limitations, obstacles, and of course, capabilities. Above all, I think that cooperating is building trust. You exchange information because you have trust. You have trust because you interact with an organization or a partner that will protect the information, that will use secure channels, and that will have a legitimate, clearly defined purpose. Interpol is not the data owner, by the way, of all of this information. This is our clients. This is the law enforcement community. This is you. I would like to summarize the next part of my intervention in three words: connect, understand and coordinate. That's the three main contributions that, again, help Member countries building trust. Let me take you through each of them. First, connect. I talked about the necessity to connect with secure channels, to have the right information, the right country, the right actor at the right moment. That's very important. International cooperation needs to achieve this reliable network. And that's what our national central bureaus are, a network, a community speaking the same language, having the same tool, and having reliable contacts to each other. NCBs are trusted national interlocutors and centers of excellence, connecting national law enforcement, all the national entities, as we name them, to this international network. When you receive massive amount of information every day, you cannot miss the information that matters. That's one of our challenge. So still decades, Interpol provides secure channel, which is called I-24/7, you may have heard about it, to exchange security information. We are progressively moving towards a new messaging system, smartest, with AI features as well, that is called Nexus. The idea is to be faster, smarter, and more intuitive for the end users. Because an NCB, again, that receives hundreds or thousands of messages every week, with vital message that may prevent a crime, arrest a high profile criminal, or intercept a cocaine shipment, he cannot or she cannot simply miss it. So it refers to technology. Technology must reach the officer where the officer needs it. That's another important aspect. Let's imagine a border, not an airport, not a place where you have all the comfort necessary to conduct checks, but imagine a border in the desert. a border, along a river, in the jungle. That's the reality of our police officers on a daily basis. No electricity, no Wi-Fi sometimes, no 5G. Yet, this may be exactly where a network is moving people, weapons, drugs, or stolen vehicles. That's vulnerabilities. We must fill this gap. A database at a quarter is of limited value. You need that the information travel with the officer, with the police, to be accessible. So, Data creates value only at the point of action. And that's why we have developed mobility in our capabilities, meaning that the importance is to have this, again, when it matters. So it is Interpol mobile devices. You can bring a sample of our databases, not the 90 19 databases, of course, but a sample that will be useful for you based on your role on the field to conduct your checks at land, air, and sea borders. Let me illustrate with one example. That's an operation that we have conducted in January 2025, focusing terrorists. It's called Operation Screen. It was based in West Africa. The results speak for themselves. 1.7 million checks were conducted. resulting with 62 arrests, including nine suspected terrorists, as well as weapons, explosives, drugs, counterfeited medicines, stolen vehicles, and so on and so forth. But it's not only regular database that you should have access to when you are conducting a verification on the ground. You need also to have biometrics. That is very key for police activities. So we have developed the biometric devices that enable enrollment and live comparison of fingerprints, facial images against Interpol databases. This is, to my opinion, especially valuable in the penitentiary settings to help our member countries to identify high-risk criminals and terrorist profile that are in these facilities. But again, collecting billions of records doesn't automatically make us smarter and more efficient. It leads me to my second point, understand. the knowledge. The problem has changed. And I think that we sense that from every of our speakers' presentation. The problem is not the scarcity of information, but the abundance. We have sometimes too many information to deal with in unstructured way, to have different formats. And you need to make it clear from a cloud, from a data set, large data set, to have actionable information. And that's the role of criminal analysis, of intelligence crime, to connect these fragments and to produce a shared understanding of the threats. The threats leads to priority. The question is no longer do we have information, is can we identify the signal in the noise? And that's what we have done, or we would like to contribute to that to our member countries, is the platform, the analytical platform called Insights, that brings separate data sets that we have into one analytical view. The idea is to click once, and you can query all of our data sets. And thanks to our donor, the US Department of States, for their support on this. It was an important endeavor. The information can be brought together again and queried more efficiently. But before to achieve that, we had to, and that was also based, something that was reflected in some interventions, to have a unified information model to harmonize all the systems and to connect them and to be interoperable, to have the possibility to integrate the systems in domestic systems and technological environment. That's very key. We need also to, and that's what we are working on, to have AI boosted capabilities to help prepare and process large data sets before it enters into our system. So another question is, can we transform the signal into something people can use? And who is the public? An officer, an investigator, a decision maker. We need to address both of the operational sides, that we need to adapt their tactics on the ground to develop investigative strategies to tackle a criminal group, to find new angles and focus on the criminal profiles that matter. And on the same time, because it's really virtuous cycle. It's the strategic side. I mean, the decision maker, the policy maker to reassess the priorities with all the information necessary to allocate the resources where we don't have unlimited means and to adapt policy to what is happening on the ground. And this leads me to my Third, if you have properly followed me in my introduction to my first key word, is coordinate. Coordination is key. Coordination based on the operational work. That's also a capability of INTERPOL, meaning that we coordinate, we connect the effort on the ground, who acts, who is informed when you are working with different national agencies, which authority leads. which steps happen first and what action must happen simultaneously. And then it's not enough because this coordination then creates a feedback loop to feed us back with information that we have seen from the ground. And again, the most insightful information that I may be able to have in my career has always been from the investigation, from something that you have found during a search, something that has been said, maybe not really clearly by someone during a custody. That's a low signal that must be interpreted and turned into intelligence, into actionable information. Police data shows the manifestation of crime, not the whole ecosystem. So police data is not sufficient in itself. It may not reveal all the conditions that enable crime or the factors shaping its future trajectory. We need open source information, we need academic research, we need the expertise of the international organization, the civil society, as well as the private sector partners. These perspectives complement what law enforcement sees through operational activity. So let me finish with this aspect. We must, we have to enrich police information, to challenge it, to place it into context. And that's where partnerships help us moving from seeing events to understanding the systems, to have the big picture, as we say. from criminal activities to criminal ecosystem and ultimately from reacting to threats to anticipating them. Isn't that what we really want to stay one step ahead of crime? Thank you.

Thank you very much, William. You have started by throwing a lot of data at us, 168 million data points and than 9.2 billion searches, 300 searches per second. I don't generally get migraines, but thinking about that would give me a migraine headache. I love that one point you have of international cooperation is not an option. I think we should all remember this. Yes, it is an option, but the other option is that We simply are in great difficulties if we do not respond when we do not plan ahead as to what we should be doing. And I also take your very French tripartite presentation on connect, understand, and coordinate. Thank you very much for these points. It is now my pleasure to give the floor to this workshop's last panelist, Ms. Leticia Ferhati, who is also a member of the Generation Justice Youth Network. She is a joint LLM candidate in international criminal law at Columbia Law School and the University of Amsterdam, and currently works in international institutional legal affairs in The Hague, the Netherlands. She's previously worked at international courts, the United Nations Office of the High Commissioner for Human Rights, and the World Health Organization. You do get around, don't you? She completed her legal and international relations education in Geneva, London, and Singapore, and has undertaken specialized training in space law at the European Center for Space Law of the European Space Agency. It is my pleasure, Laetitia, to give you the floor.

Thank you very much for the presentation, distinguished delegates, colleagues, co-panelists, and chair. Workshop three asks how new data sources, analytical approaches, and policies can help us get ahead of emerging crime. Satellite and geospatial data speak directly to all three. However, while they can help us see harms that traditional reporting systems cannot reach, Their increasing use also raises a fundamental question, which is how do we ensure that this data can be trusted, challenged, and defended when it becomes legally consequential? This presentation, therefore, looks at the standards, transparency, and cooperation needed to move from simply accessing geospatial data to using it responsibly as evidence. As geospatial data becomes more central to preventing and prosecuting crimes that traditional methods cannot reach, This presentation argues that the private actors who produce it need to be brought into a framework of shared standards and transparency. In this sense, traditional crime statistics rely heavily in reports, administrative records, and surveys. But many emerging harms, from cybercrime and trafficking to crimes that affect the environment, do not fit that model. This is why the Congress concept note points towards incident-based, near real-time sources, and satellite imagery and sensors. Two sources are particularly relevant in this context. Citizens can provide contemporary news accounts from the ground, while satellites can provide geographically referenced information where physical access is difficult or unsafe. Neither is necessarily sufficient on its own. A witness account can provide context and immediacy, while satellite imagery can show where changes occurred and how they developed over time. Together, they can help establish timelines, locations, and patterns, allowing one source to corroborate or contextualize the other. This is particularly valuable where traditional investigative methods are constrained. Geospatial data does not replace human evidence, and it can extend the evidentiary picture beyond what investigators can observe directly. This is not entirely new territory for the United Nations. Remote sensing is already used across UN mandates, including for environmental monitoring, disaster response, and documenting human rights situations. The question for criminal justice is therefore how to develop shared standards through combining, verifying, and ultimately relying on these sources of evidence. States and regional organizations already use commercial and public earth observation platforms to document destruction, demonstrate changes over time, and corroborate witness accounts, as well as monitoring activities ranging from illegal logging and mining to maritime activities. For example, satellite imagery can identify areas of illegal deforestation, while satellites and other monitoring systems can track vessels and detect suspicious acoustic sensors or AI-based analysis to produce more timely intelligence. These examples show that the technology is already integrated into operational decision making. The challenge is no longer simply access to the technology, but establishing its reliability of the, when it becomes evidence. If an image is later relied upon in an investigation, we may need to establish its acquisition date, source, sensor characteristics, processing history, and analytical methodology, as well as how it is combined with other data sets. The image may be accessible, but the process that produced and interpreted it may not be equally transparent. And we already are seeing this issue in international justice. UNITAR and UNOSAT have developed dedicated work on the technical and legal requirements for using satellite imagery as evidence in international judicial proceedings, reflecting the growing role of such imagery in investigations and court room proceedings as such. There is, however, a legal gap. Space law establishes who is responsible for space activities, who is liable for damage caused by space objects, and which state retains jurisdiction and control over a registered object, principally through the Outer Space Treaty and the 1972 Liability Convention and Registration Convention. Criminal justice asks different questions. When satellite imagery is used as evidence, we need to establish how it was acquired, processed, and interpreted, and whether its origin and integrity can be demonstrated. The legal significance, therefore, extends beyond the satellite itself to the information it produces and the evidentiary process through which that information is used to establish facts. Space law governs the object and the responsibilities surrounding it. Criminal justice must scrutinize the evidentiary life of the information generated by it. Why is standards arrive too late? There is also timing problem. Investigators act first, while courts may assess the evidence years later. If tasking records, metadata, or processing history are not preserved when the evidence is created, a later court may not be able to reconstruct them. Standards, therefore, have to exist before the evidence reaches the courtroom, which is not simply a question of admissibility, but of designing the evidentiary process from the moment the data is required. Furthermore, one important principle is corroboration. Where independent sources, for example, optical and radio imagery, agree, confidence in the evidence can increase. Satellite information can also corroborate witness testimony or other investigative findings. UNOSAT's work with international justice mechanisms illustrates this practical value, and its geospatial analysis has supported international criminal justice, including in locating damage to cultural heritage sites. But collaboration also requires contestability, the ability to examine the sensor, operator, date, resolution, and processing behind the evidence. And satellite data also has clear limits. It can show that a physical change occurred without necessarily explaining why and who was responsible and needs to be held accountable for it. Therefore, other sources and appropriate expertise remain indispensable in this process. This is the basis of what this presentation outlines as an orbital evidence doctrine built around three stages, acquisition, processing, and interpretation. Each introduces different risks. An authentic image can still be processed opaquely. Sound processing can still support a flawed interpretation. And a technically accurate interpretation may still be insufficient to establish responsibility without corroborating evidence. Evidentiary standards should therefore correspond to the consequences attached to information. Monitoring and due diligence require something different from a finding of responsibility or criminal culpability. The point is not to create a single threshold for reuse of satellite data, but to ensure that the level of scrutiny is proportionate to the legal consequence. Which brings us to a transparency gap. Existing space law does not address the full evidentiary life cycle of Earth observation data, but it establishes responsibility for space activities and objects, and not how imagery should be preserved, processed, or interpreted later on. International practice shows that shared non-binding standards are possible. The Berkeley Protocol, for example, provides guidance on collecting, preserving, and analyzing digital information for investigations. What is still missing is a comparable layer for commercial earth observation evidence. If a private operator provides imagery showing destruction, for example, an investigation may need to establish when and how it was acquired, what processing was applied, what other data were used, and whether the resulting analysis can be independently challenged. As more evidence is generated outside government, the question is therefore not simply whether the data can be accessed, but whether its provenance and interpretation can be traced and contested when it becomes legally consequential. This presentation therefore suggests three steps. First, standardizing acquisition, processing, and interpretation so that basic information about how evidence was produced is consistently recorded. Second, preserve contestability so that those affected can meaningfully scrutinize and challenge the evidence where appropriate. And third, building an inclusive institutional forum, bringing together states, statisticians, investigators, satellite specialists, and affected communities. Satellite data can help us get ahead of emerging crime, particularly where traditional evidence cannot reach. But its value will ultimately depend on whether we can trust it and defend its use. The task before us is therefore not simply to expand access to data, but to build the standards and cooperation that make that data legally and operationally credible. Thank you for your attention.

Thank you very much, Leticia. You brought us into the space age with this gathering of data and pointed out the not only the potential for it, but also the difficulties that lie in ahead of us when we try to use this data, much of which is produced by private operators. I must admit that I myself have been surprised by the fact that not only are more satellites being launched, it's about every week, but more and more of these satellites are being operated by private operators. And then you had closed in your last slide with the suggestions for where we should go from here on in. I'd like to thank the three speakers for our second session, and I would first like to ask if any of the panelists, either from the first session or the second session, would have questions or comments regarding these last three presentations. That doesn't seem the case. Oh, yes, please, Ari.

Thank you, dear Matti. A quick takeaway, perhaps, so I don't, perhaps something of a comment, especially pertaining to what Mr. Williams so kindly put forth at the end of your presentation, and perhaps to summarize some of the key findings from yesterday. We speak about data on a fairly technical level a lot of times, obviously, but at the same time, I think something I find intriguing is the concept and necessity to kind of harmonize different sources data, private data has been mentioned as well as the civil sector data and kind of a variety of different sources to then reach conclusions on what we actually conceive and perceive as threats, both operationally but also to some extent conceptually, what are we going towards? Obviously, we have the legal frameworks, the national legal frameworks and of course from international law as to what constitutes crime and what we prioritize like the esteemed delegate previously put forth in another question, but perhaps something of a takeaway. I think we need a meta discussion, perhaps, on how to further utilize these other sources of data in harmony with perhaps more official crime statistics type of data and so forth. I don't know, any takes on this?

Thank you for that comment, Ari. William, would you like to respond?

Sure, thank you very much for this input. I think that, yes, that's the very point of international cooperation, is to deal with lots of different types of information and dealing with different frameworks. So that's where Interpol also is ensuring a dedicated framework to make this happen, meaning that the idea is to at least that all the contributors agree upon a common understanding about how the data will be used, where it will be ingested, and what will be the final product that it will generate. Once these answers are providing, then it creates the trust that I've mentioned, and it makes the overall cooperation more efficient. But it comes with lots of challenges, especially when we see how the technology evolves and the need now that we are able to harmonize digital formats. Because finally, data is not more than a digital information. But you have to deal first with this digital aspect, and then you deal with information.

William, if I could follow up on this one. Of course, I have myself served as a judge, but I have been struck by the fact that law enforcement agencies in general, even domestically, they tend to be, shall we say, rather protective of their data sources. And you have been working in an environment where, as you yourself said, almost 200 that law enforcement agencies of over 200 nations are required basically to cooperate and without their cooperation, without the trust that they have in Interpol, it would be very difficult indeed. Could I ask you very generally about how has Interpol tried to encourage or shall we say garner this trust and has there been any shift in this, has it become more difficult or less difficult? My own understanding is that it would become less difficult because more and more law enforcement agencies domestically are beginning to realize, oh gosh, we do have to have this cooperation. As you yourself said, there is no option to that.

Yes, thank you very much. Again, it's a very important challenge. I think that if you share information, You need to know why. You need to have a purpose. And this is to obtain something additional as an intelligence or an investigative lead. So once you have this purpose, then you need to ensure that the data that you contribute to Interpol is protected. And for this, it's not to, of course, close this information into a sort of impenetrable shell, otherwise nobody will know. But it's to make sure that us, as Interpol staff, we are able to deal with the data and to respect the restrictions to make the connections, what the connections are made. Then this is our job to ensure that the different stakeholders agrees upon how this can be used ultimately within the proceedings or for intelligence reports. We have different frameworks. We have the general frameworks that applies general restrictions. So you can decide that you would like to share with everyone, that's the notices, for instance. If you send the notices by design, it will be disseminated to all of our member countries. But then you have the diffusion. The diffusion is sort of more restricted format for the information that you would like to share. We have also specific framework called the criminal and ethical files. We talked about the necessity to connect the data and these criminal ethical files are temporary basis, thematic, and only with participating member countries that will accept by exception to have a larger opening of that data, accessibility of the data to the analysts, and then to provide analysis in return. And that's the very point of having created recently the Global Analysis and Intelligence Network, which is a sort, I would say, crime analyst coalition working not only on the data, but also on what we've learned from the data. Thank you.

If there are no further comments from panelists, perhaps, Mr. Chairman, if I could turn the floor over to you to see if the audience would like to comment.

Yes, exactly. Thank you. And I see the United States again is ready to break the ice. The floor is yours.

Thank you, Chair, for indulging yet another U.S. intervention, and thanks to the panelists again for great presentations. I'd like to comment on our panelists from Interpol who raised some really important non-numeric types of data that are often underused to dismantle criminal networks, and that's biometrics and child sexual exploitation and abuse data. And so we know that the U.S. is a leading advocate for sharing biometrics information in 2025 through a CCPCJ resolution. States committed to share biometric information sharing. It's the first resolution outside the U.N. Security Council calling on the sharing of this type of information, which we see as pivotal to take down alien smuggling networks and to strengthen border security. but also child sexual exploitation and abuse material. There are thousands and thousands, perhaps you could give us the exact number of unidentified children in Interpol databases. and the US through the Homeland Security Investigations, HSI, partners with Interpol and others every year in an operation called Operation Renewed Hope, where we bring investigators and analysts from various jurisdictions to come together to not only share analytical tools, but also to leverage different legal authorities, specifically some that we have in the US that allows us to use facial recognition technology and other types of technology consistent with US law to identify some of these child victims and to rescue them, and not only the victims, but the criminals themselves. And so we just call attention to these types of data and would have a couple of recommendations in that regard. Number one, states should improve international information sharing and cooperation, including one, sharing biometric information sharing to disrupt alien smuggling networks, and two, sharing analytical tools and expertise to identify child sexual exploitation and abuse victims and perpetrators. And with regard to the last presentation on geospatial data, it's also fascinating because I think it It answers the question that data in itself really amounts to nothing if there's no action behind it. And so very curious, the idea of scam centers, for instance, in Southeast Asia, we know where they are. They're on a map. Everybody could point to exactly where they are. They're industrialized scam centers. They're enormous. There are more and more that are growing. And so it's not so much a matter of whether or not we have the data, it's a matter of whether or not we have the political will to dismantle those networks when we see them very clearly on a map. And so our last recommendation would be to really urge states to make use of this type of data, including geospatial data, to actually take down these networks, including scam centers. Thanks.

Thank you. I saw a lot of nodding here on the podium. I don't know if somebody wants to react directly, you know, then we Yeah.

Thank you. Me again. Thank you very much, dear colleagues, for your intervention on two strategic topics, two key topics. First, biometric. Just to set the scene regarding biometrics, Interpol launched three years ago now already the biometric hub that provide a platform for all the member countries to connect their domestic biometric system, related to fingerprints and facial recognition. The DNA is a separate database. So it means that now you have the possibility, first, to exchange information to make comparison, two, to contribute to the database in view of potential further connection, and three, to integrate fully the biometric app to make the connection instantly. That's the first aspect, and I'm fully aligned with the fact that biometric is very, very key for most of the major activities that we that we face. So, I expect is yes, the child sexual exploitation material. You may know that we have the ICSU databases, that database, which is the international child sexual exploitation databases. We are launching a new phase to make it more adapted to first identify victim because it's not only identifying perpetrators, it's also to protect victim and to enlarge the possibility in terms of compression because the technology is advancing. So the next generation is also a big topic that we are currently developing in a top. Thank you.

Thank you. Are there more requests for the floor? Questions, comments. If this is not the case, then I'd like to ask Marty to summarize the insights of today.

Thank you, Mr. Chairman. Ladies and gentlemen, this is, in my opinion, what a UN Crime Congress workshop should be like, much more interactive. I found this a very dynamic event. And I might also reveal to those of you who don't know that even though it seems there are very few participants here in the room, quite a number of people have been following our discussions online. So it has generated quite amount of information. This has been a technical workshop, but it has provided lots of examples from member states of the United Nations and from various organizations as to what the challenges are, but even more important, what the possibilities are for getting that data that would help us in our decision making. Several takeaways from this, I may in part be repeating what I said this morning when we began, and also what my colleague Buki said in closing yesterday's session, but Crime data exists. The data we need to respond to our present crime problem, however we want to define that, but also the problems that will emerge over the years to come, it exists. It is out there, but it's scattered in different sources. It takes different forms. Various different indicators are done, and I note what William said towards the end of his presentation that one of the dangers is that there is too much information out there and we have to sift through this information to find out what is, if I may use a term used by Salome, actionable data, or William, actionable information. So what is that information that we need which can help us provide a better basis for our policymakers and more directly in operational activities for our practitioners. We do need to bring these heterogeneous sources together. Ari, you had referred to harmonizing these sources of information. I think it'd be more a question of bringing them together, harnessing these different sources of information, because they will remain different. simply because they arise from different sources and they do take different formats. We need to produce those more reliable indicators as soon as, and I take the point of the distinguished representative of the United States, we have to identify what the priorities are so we know where we should target our information gathering activities. We also need to look beyond our present crime data into the future using the foresight skills referred to by Ari in the first presentation this morning. But we are learning. I find this is very encouraging. We have the example shown by our Korean colleagues of how information can indeed be harnessed, can be formatted, and can be put into a form that to some extent, open data, which can be used more generally, not only within decision making by government authorities but by academia, researchers, organisations and the private sector in their own activities. When we start talking about open data, we're talking about data on crime in general. This raises issues of data privacy, data reliability, et cetera. These are issues-- and data security. These are issues that do need attention domestically, developing standards that apply domestically, but we also need cooperation in this. so that we can have a basis for international decision making. This implies, and this I will refer to what Ottavia from UNICRI had said, and Salome for the UNODC had said, a need for some kinds of common understandings for consistency, for comparability, and for interoperability. We are dealing with 190 member states, we are dealing with legions of private sector agencies, organizations, et cetera, but we do need to have, to some extent, a common understanding of how this information should be processed and used. I also note that there was a reference to the fact that there is a different maturity in different states when it comes to developing or harnessing data in decision making. This brings up the issue of cross-border cooperation, not just the operational cooperation that comes in the international financial flows referred to by ISRA, but also the type of operational cooperation that is institutionally represented here by INTERPOL. We also need technical cooperation in helping member states and private sector agencies around the world in developing their data and getting that integrated into decision making. William did have that comment that international cooperation is not an option, emphasized the importance of trust, but I think through discussions like the ones that we have had at this workshop, it shows that there is a recognition of the need for international cooperation. Now we simply have to take that one forward and work together in an atmosphere of trust for better cooperation. And on that point, Mr. Chairman, I close. I wasn't sure if it was you, Mr. Chairman, or myself. But I would like to take one last stand as the moderator, and that is to invite the director of the Thailand Institute of Justice, who has been the one who has organized the work on putting this workshop together in very close cooperation with the UNODC. Director Phisit, you are there. May I ask you to come perhaps to the speaker's lectern?

I can maybe stay up here, if that's okay. Thank you so much, Matti. Mr. Chair, with your kind permission, I would like to share a few observations on behalf of the Thailand Institute of Justice. has had the honor of organizing this workshop together with the UNODC. Of course, this is the first time that the THA has led an official workshop at its crime congress, and it has been a privilege for our team to work with the UNODC, our PNI partners, and all those who have contributed to the discussions over these past two days. Of course, the subject of this workshop, strengthening data collection and analysis, is not new, but it has become increasingly important as crime continues to evolve, often in ways that challenge our existing methods of detection and responses. Please allow me to maybe share a few brief observations from our discussions about the model has already summarized nicely, but I think from my point of view, I think We have seen the complexity of new and evolving forms of crimes, and this reminds us that no single institution or even source of data can provide a complete picture. We need to make better use of information held by different actors, both within and beyond the criminal justice system. Of course, innovation offers important opportunities to improve how we collect and analyze data. but we also need to really come up with standardization and collective effort in order to make better use of this technology. And third, I think the most important point that I gather is that better data should ultimately help us make better decisions. And I think in this regard, countries have different capacities, face different challenges, But we do share the need for reliable information that can inform crime prevention and justice policies. And for this THA, we continue to explore the important issue of how we can translate the knowledge and experience shared at this workshop into practical approaches that we can help and support national and regional cooperation while keeping people at the center of our efforts. So on behalf of the Thailand Institute of Justice, I would like to express my sincere appreciation to the UNODC, our PNI partners, the speakers, the moderators, all the participants, including member states who have shared interest and also have contributed to this workshop. TIJ look forward to continuing our cooperation and building on the ideas shared in this workshop. Thank you so much.

Yes, thank you also from my side to all panelists and speakers and the audience. Distinguished delegates, ladies and gentlemen, we have now reached the conclusion of the workshop three. Let me inform you that the report of workshop three will be brought to the attention of committee one on Wednesday, 30th of September. in the afternoon and adopted by the plenary in the morning of Thursday, the 1st of October, according to the proposed organization of work for the Congress. The report will consist of the proceedings as well as the summary by the Chair, containing a summary of the deliberations and Chair's conclusions and recommendations drawn from the discussions. In preparation, I will rely on the assistance of the Secretariat. and on the summaries of salient points of the discussions made after each panel by the moderator of the workshop. I will also rely, and I'm thankful for this, on the support of the rapporteur of the workshop in ensuring the precision and clarity of what is to be reflected in the report. And as mentioned before, Mrs. Tanya Wyatt will make a short presentation on salient points raised during this workshop at the opening of agenda item five in the plenary. Yeah, again, thank you very much for your participation and my special appreciation to our moderators of the workshop. And we will resume work in committee one on Wednesday afternoon for the adoption of our report. And with this, the meeting is adjourned. Thank you.