For more than 30 days, it endured active fighting and life under occupation. It was liberated in early April, but the scars of war remained. At the local gymnasium, shells that landed nearby shattered the windows and damaged the sports hall. Because of the destruction and the constant danger of repeated attacks, children had to switch to online learning.Later, with funding from the European Union, UNOPS repaired the gymnasium building, allowing the school to reopen its doors. However, another challenge remained: although the old basement provided shelter during air raid alerts, it was not suitable for prolonged stays. To restore comfort and safety for children, UNOPS later fully renovated and equipped the basement. It now has everything needed to ensure the uninterrupted continuation of learning.The renovation was completed in mid-May 2026, allowing students to finish the school year in person.    “I’m glad we now have a shelter. We’ve already gotten used to it: whenever the air raid siren sounds, we immediately go downstairs and continue our lesson there. Every class knows where to go, so everything happens very quickly,” says Illia, an eighth-grade student at the gymnasium. Even now, during the summer holidays, children come to school for additional subject consultations, and they sometimes have to go down to the shelter, as air raid alerts are frequent in the region.“After the ninth grade, I plan to enroll in college. I understand how important it is to attend school now so I can prepare better. Sometimes a lot of drones fly over our village, but when I’m at school, I know I’m safe in the shelter. And it gives my parents peace of mind too,” says Sofiia, a ninth-grade student. For the gymnasium, the renovated shelter became much more than a technical solution. It made it possible for the school to remain a school—a place where friends come together, where life goes on, and where learning never stops.   About the projectThe Schools Repairs in Ukraine project, funded by the European Union and implemented by UNOPS, aims to repair educational facilities that have been damaged during the ongoing war in Ukraine. This $30 million initiative is being implemented in the regions of Chernihiv, Kyiv, Sumy and Kharkiv, focusing on schools requiring light to medium repair work.The primary objective of the project is to provide access to repaired school facilities, which will increase the proportion of children receiving face-to-face education. By rehabilitating more than 70 schools, the project will help restore facilities to a safe and welcoming environment where children can continue their education without interruption.

Mine action is not as simple as "mines bad, clearing mines good." Every hectare we clear, and every hectare we choose not to clear, carries an environmental cost, and pretending otherwise does a disservice to the donors, governments and communities we serve. If we want mine action to take the environment seriously, we first have to be honest about the trade-offs, and about what taking it seriously would actually cost.A Sector Touching Every Part of RecoveryMine action touches every aspect of life in a mine-affected country. You cannot rebuild a school, build a hospital, drive down a road or plough a field if there is a concern about mines. It is also, whether we have treated it that way or not, an environmental issue.In Ukraine, one proposal has been to issue a green bond to put wind farms into minefields: clear a small area for a turbine at minimal cost, let the turbine generate power and revenue, and use that use that revenue to fund clearance of the rest of the field one turbine at a time. It is a genuinely green project, with  sustainable energy paying for mined land to be cleared while building a independent green source of sustainable energy.At the other end of the scale, when the Kakhovka Dam was destroyed on 6 June 2023, in what Ukrainian officials at the time described as an environmental catastrophe, the collapse released roughly 18 cubic kilometres of water from the reservoir in a matter of days, flooding towns and farmland downstream along the Dnipro River and displacing thousands of people. The rushing water tore through riverside positions and former front-line areas, dislodging landmines and unexploded ordnance and carrying them, along with fuel, agricultural chemicals and military waste, into areas that had never been mapped or cleared. The environmental damage it caused, toxic chemicals and debris swept downstream, became far harder to address precisely because the areas affected were either already contaminated, or suspected of contamination, with mines: teams could not safely reach flooded land to assess or remediate it until it had also been checked for explosive hazards. Explosive ordnance itself, particularly the more complex items, over time leaches chemicals into the ground so for as long as it sits there contamination is occurring. Even our standard method of destroying old ammunition, open burning, means putting toxic chemicals into the atmosphere to remove a hazard from the ground. The Case for Restraint: Zone "Rouge" and the DMZHowever not all contaminated land needs to be cleared, and some of it may be better left alone. After the First World War, France declared parts of its former battlefields a Zone "Rouge", too dangerous or too costly to clear, and left them. At the time, that zone covered more than 1,000 square kilometres; today it has shrunk to roughly 100 sq km, and that remaining land is now among the most biodiverse in the country. The Korean DMZ tells the same story: decades free of human disturbance, because it is mined, have made it one of the most biodiverse strips of land anywhere in the world. Clearing the mines will almost certainly impact this balance. That is not an argument for abandoning clearance. It is an argument for asking, area by area, whether leaving land uncleared would in fact produce a better environmental outcome than clearing it, and building that question into how we prioritise. In Ukraine, a tool like GRIT, a national digital platform developed by Ukraine's Center for Humanitarian Demining and Ministry of Economy, Environment and Agriculture with UNDP support, could in principle be extended for this purpose. GRIT already draws on more than thirty layers of date from state and commercial datasets, including minefield records, soil & forestry data, satellite imagery & population movement, using AI/ML models to rank clearance priorities by their impact on people, agriculture, infrastructure and economic recovery. Extending that same logic this same data could be used to flag which areas make sense as long-term "red zones", rather than absorbing scarce clearance resources for marginal gain, would be a natural next step.The Damage We Do While Trying to HelpDemining machines, flails, tillers and combinations of the two, dramatically improve the pace of clearance. They also pulverise the ground, destroy root systems and soil structure, release stored carbon and can permanently alter the biomass of a site. I have seen fertile land in Afghanistan effectively destroyed this way, when topsoil loosened by a machine was blown down an entire valley by the wind.None of this means we should stop using machines. It means machine use needs proper planning around soil type, season and weather, and it means the sector should be thinking harder about remediation: something as simple as fertiliser, seed and fast-growing cover crops sprayed behind a machine to help soil recover, rather than leaving pulverised ground exposed. We do not yet have good answers here. We should be looking for them. The Missing Fourth Dimension: TimeMine action is not static, certainly the environmental risk and benefit from it is not and treating it as a single snapshot in time misses the point. Some contamination is leaching into soil or water right now and needs an urgent response. Other munitions will not degrade for another thirty years, some will never cause significant contamination, which raises a genuine question: is it better to clear a complex, with resulting high-impact to the environment site today, or to wait ten years for demining technology, very possibly including autonomous systems, to do the job with a fraction of the environmental and human cost? The same logic applies to human risk. Every task assigned to a deminer carries both a financial cost and a real chance that the person doing it will be killed or lose a limb. Weighed against a five-year wait for safer, cheaper, autonomous alternatives, that is not a comfortable calculation, but it is one we owe it to our operators to make it honestly rather than by default.Climate Change Is Already Changing the Risk PictureExtreme heat and extreme temperature swings accelerate the degradation of explosive items and can cause some explosives to separate into more unstable compounds. But the larger risk is movement: flooding and landslides shift contamination from known, marked and understood areas into places nobody is expecting it. In Ukraine, where light plastic items like PFM-1 "butterfly" mines are present in large numbers, floodwater can carry them well beyond any mapped hazard area, as we saw with the Kakhovka Dam. That is arguably the single biggest climate-linked risk mine action now faces: not that hazards get more dangerous where they are, but that we lose track of where they are at all. Vision, Then Resourcing, Not the Other Way AroundNone of the above is worth much unless the sector is honest about cost. If the United Nations' mine action work decides that environmentally conscious programming should be standard practice, that is a legitimate choice, but it is not a free one. Ten per cent of a programme manager's time redirected to environmental considerations, multiplied across twenty-odd country programmes, is not a rounding error; it is real money, on the order of hundreds of thousands to a million dollars a year. A modest upfront investment in better data and planning tools could, over time, cut that ongoing cost substantially, but someone still has to choose to make that investment, and someone has to tell donors, clearly, that a higher environmental standard comes with a higher price tag.We tried a version of this a decade ago with an initiative called Greening in the Red. It produced good intentions and very little else, because it was never built into strategy or given a budget. The lesson was simple: a memo asking people to try harder is not a policy. Not a Simple StoryI say this as someone who has spent thirty years in this sector: nothing here is as simple as "mines bad, clearing mines good." A fenced, marked anti-personnel mine on a former front line poses little risk to anyone today. The same mine, still active in twenty years once the fencing has gone, the trenches are filled  and the front line is a memory, is a very different problem. But again a handful of scattered mines on land that will never be used again may do less environmental harm sitting in the ground than the fuel burned and the machinery deployed to remove them. If mine action is serious about the environment, that seriousness has to show up in three places: in the vision the mine action sector in Ukraine sets for itself, in the strategy that follows from it, and in the resources allocated to deliver it. Anything short of that is a good intention without a plan to deliver a clear outcome, and in this sector, as in most others, good intentions without a solid justification and a plan tend not to survive contact with a budget review committee. 

By Vitalii Zakharchuk, UN Ukraine RCO Senior National PSEA Coordinator.Sixteen years in civil society and twelve in humanitarian work have taught me one thing: whenever a society treats vulnerable groups as a burden, it loses. People with disabilities, veterans, internally displaced people, LGBTQI+ community members — all of them usually get filed under “aid recipients.” A donor requires that they be “included,” a service organization “focuses on their needs,” and the system keeps treating this work as a cost line rather than an investment.For years I have explained the opposite to colleagues using Einstein’s school formula. E equals m times c squared. E is energy, the outcome: a strong, resilient community where everyone can reach their potential. m is mass — in my analogy, that means people, every resident of a community, including those we label vulnerable. In physics, even a gram of matter hides enormous energy. In society, every person hides potential that usually stays invisible only because of the barriers around them. And c squared is the accelerator: accessible infrastructure, absence of discrimination, trust, equal opportunity. That is what turns a community’s passive mass into active energy for development.In practice, it looks like this. A traditional approach sees a person with a disability as a budget line. An inclusive approach understands that removing a barrier turns that same person — from a benefit recipient — into a taxpayer, a volunteer, an idea generator, an active citizen. A ramp or a low-floor bus is not only for a wheelchair user. Mothers with strollers, older people, injured passengers, couriers with heavy bags all use it too. One accessibility feature raises mobility for several groups at once, and that is the entire point of the c² accelerator.There is another effect people rarely mention: social synergy. In physics, splitting a nucleus triggers a chain reaction. In a community, integrating different people creates new connections. When business, government, activists, and people with different lived experience sit at the same decision-making table, unconventional and genuinely workable solutions emerge. That is not a slogan — it is something I have watched happen while coordinating the PSEA network in Ukraine.The question I get most often is some version of: is inclusion even possible right now, when there are fewer resources, less time, less safety? My answer stays the same. Inclusion does not compete with a community’s survival for the same resource — it is how that resource stops being lost. When a community excludes part of its people from decision-making, it simply writes off part of the mass m in its own formula, then wonders why there is less energy than there could have been. War does not cancel this arithmetic. It only raises the cost of the mistake: a person who was not heard before 2022 is now much harder to find, because they may have fled, lost their documents, or ended up under occupation.Formal inclusion, and what it fails to seeWar does not create the problem of formal inclusion. It exposes it.Ukraine has building codes meant to make premises accessible to people with disabilities. On paper, everything looks flawless. In practice, I regularly see a ramp built at an angle no one can climb unassisted, or an entrance that technically meets the code but with doors so heavy, because of that same code, that a wheelchair user cannot open them alone.Here is another example of how formal compliance stays blind to a person’s actual experience. Rooms for feeding and changing infants are traditionally placed only inside women’s restrooms. A father holding a baby simply has no such room available to him. Inclusion exists in the label on the door, not in the logic behind it.Sometimes it works the other way around. In Lviv, and later in the office of a humanitarian organization in Kyiv, I saw restrooms made gender-neutral and, due to the building’s own technical limits, wheelchair-accessible at the same time. They had also placed a set of period and hygiene supplies there. No loud statements about inclusivity, no report about “vulnerable group engagement.” Just a decision made by people who thought about how a real person would use that space, not an abstract beneficiary from a project proposal.The difference between these examples is the actual test of whether inclusion works. A standard in a document guarantees nothing. What matters is whether anyone actually walked the path of the person the standard was written for.The same holds for wartime humanitarian programming. A heating point can meet every Sphere standard and still be unreachable for a wheelchair user if the only entrance has a 15-centimeter step. A payment notice can be translated into several languages and still be useless to a person with a hearing impairment if the only channel is a loudspeaker in the square. Formal compliance and real accessibility are not the same thing, and the gap between them is visible only to someone who checks the outcome itself, not the report about it.When inclusion becomes a matter of dignity, not paperworkTwo years ago, I had surgery, and during my rehabilitation my abdominal muscles were supported only by a corset for a long time. I started getting around Kyiv in that condition and realized, for the first time, just how painful and difficult it was — not in an abstract sense, but physically. On the subway, passersby who were simply in a hurry would bump into my stomach. Once, an event was held on the “minus one” floor of a hotel, and I had to navigate the steep stairs leading there on my own. Offices, hotels, public venues, and institutions all proclaim inclusivity on their websites. But almost none of them put this into practice, because none of the people writing these statements had any idea what it means to move around when something in your body is not working the way it should.That is where inclusion stops being a word in a project report. It becomes a question of whether a person gets through the day without pain and humiliation.SEAH: why vulnerability is always about powerSEAH — sexual exploitation, abuse, and harassment — in a humanitarian context is, above all, about a balance of power. Not in the military sense, but in the sense of authority and dependency. The more vulnerable a person or group is, the greater the power asymmetry becomes between them and whoever controls the resources.That dependency has a direct, almost mechanical logic. A person who has lost their home, belongings, and documents sees a humanitarian worker as the sole “gatekeeper” to basic goods: food, shelter, cash assistance. The fewer alternative lifelines a person has, the stronger that dependency, and the fewer levers they have to defend their rights if something goes wrong.That is why the “Leave No One Behind” principle, inside the UN system’s presence in Ukraine, is not a declaration but an integration rule underpinning humanitarian response and recovery programming.The same logic runs through EU integration programming. But a rule on paper and a rule in practice — this is, once again, the gap between form and substance described above.I have an exercise I sometimes use in SOGIESC trainings: imagine waking up in a different body, like in a body-swap film. What do you feel in the first few seconds? The question sounds naive, but it exposes the most common mistake in the sector — decisions about vulnerable groups are usually made by people who understand their problems in a stereotypical way, not from the inside, behind closed doors, without a single representative of the community being discussed.The line between genuine participation and tokenism runs through authority, not rhetoric. A box-ticking consultation looks like this: an organization gets invited to a meeting, gets listened to, gets thanked, gets photographed for the report — and the program keeps running exactly as it was designed before that conversation. Sometimes it is even honestly labeled a “stakeholder consultation” in a project’s logical framework, and the community-engagement box is formally checked. But ask the meeting’s participants a month later what actually changed in the program because of what they said, and there is often no answer.Genuine participation looks different. A Community-Based Organization can step in and change a program that is already running if it sees that program causing harm or simply failing, even if that program is already funded with an approved budget. It can influence a funding decision in favor of one direction over another, not just a wording choice in a project description. The difference is simple: in the first case, a Community-Based Organization gets consulted; in the second, it gets a veto or a real vote where previously only the donor and the implementing agency decided. If nothing structurally changes after a meeting with a community organization, that was not participation — it was a show.Activism deserves a retirement plan tooOne point I keep pressing in conversations with colleagues: the humanitarian sector is used to inviting the same people to meetings, consultations, panels. The same experts, the same faces since 2014 and 2015. We do this without any bad intent, simply because it is easier and safer — vetted contacts, a familiar level of English, a familiar way of presenting.But activism, like any other field, needs a natural generational turnover. Look at how many genuine civil-society leaders have actually grown over the past decade, how many of them we helped grow ourselves, and how many we simply used year after year without investing in whoever comes next. Between 2005 and 2015, institutions invested in specific people, and those people became the backbone of the civil society we work with today. The question is who replaces them if we keep turning exclusively to leaders who are already established.“Growing new leaders” operates on two levels at once, and they are connected. It means leaders from within vulnerable communities themselves — a person with a disability or an LGBTQI+ community member who, over time, becomes the voice of their own group inside the network rather than remaining an object of someone else’s representation. And it means the next generation of PSEA and gender specialists more broadly, people who arrive with different experience and different expectations of the work.The new generation of activists and volunteers differs from the previous one, and here I will admit my own bias: they are not willing to spend their personal capacity on administration and micromanagement the way we got used to doing. They want to put a hundred percent of their energy into generating new products and solutions, not into reporting for reporting’s sake. One could read this as a weakness. I think it is worth reading as a different, and often more effective, allocation of resource — one the system simply needs to learn to accommodate.The two levels of leadership development feed each other quite literally, not just in theory. When someone from the LGBTQI+ community joins the network having walked the path from beneficiary to advocacy specialist themselves, they bring something no external training can provide: precise knowledge of exactly where the system breaks for people with their experience. And conversely, when a young PSEA specialist builds their expertise not from a textbook case study but from direct dialogue with such leaders, their professional growth happens faster and more honestly than mine did twelve years ago.What actually gets in the way, and why it is not moneyAsk what the biggest structural obstacle is, and the easiest answer is: Community-Based Organizations funding, UN bureaucracy, wartime security constraints. All of these are real barriers. But the hardest thing to shift is not money — it is the system’s willingness to change and to compromise.It matters not to confuse compromise with defeat. There is almost always a strategy where both sides win, where the international organization does not lose control over quality and accountability, and the local initiative does not lose its own identity and pace of work. But that first requires admitting that the current power distribution between a donor and a local organization is not neutral by default.The practical side of that compromise is usually simple, and at the same time, uncomfortable for a large system. Shrink a reporting template to the size a CBO can actually fill in without a dedicated administrator on staff. Accept a less polished report submitted on time over a flawless one submitted two months late. Give an organization with a team of three the same voice at a coordination meeting as an agency with a hundred staff. None of these changes require additional funding, only a willingness to redistribute what already exists.One piece of advice, when a donor or agency asks where to start: put on the shoes and walk your beneficiary’s path. Not figuratively, literally — reach the distribution point by the same route a person with a disability or a mother with a child would take. Fill out the same feedback form someone without internet access would have to fill out. Stand in the line that gets described in reports as “a minor service delay.”That alone removes half the questions about why a program does not work the way it was designed to on paper.A network the abuser will fearPeople sometimes ask how I picture the PSEA Network Ukraine three to five years from now, if the approach of inclusion plus genuine Community-Based Organizations leadership actually takes hold. My answer is fairly specific.It is a network controlled by local organizations, not an international structure that merely consults them. It will have the backing of the international community, but weight and a mandate strong enough that every person in Ukraine understands: if you turn to this network, you will genuinely be protected. And an abuser, whoever they are, understands that a real consequence awaits them, not a formal letter asking them to explain themselves.This is not a matter of one training or one memorandum of cooperation. It is a matter of whether the system is willing to hand over part of its power to the people currently on the other side of the asymmetry described above.The E=mc² formula works here exactly as it does everywhere else: energy does not appear when we talk about people as mass in a report. It appears when we remove the barrier between their potential and the chance to realize it.None of this starts with a strategy document. It starts with whoever reads this deciding to walk one path they have never walked before, sit at one table they have never been invited to, or shrink one form they have never questioned. The formula only holds if someone actually applies it.

On 6-7 July 2026, diplomats, experts, officials, representatives of civil societies and the UN agencies filled up the rooms, connected by one question: “How can we make AI a force for dignity, opportunity, and progress?”Future of risks and possibilitiesOpening the first Global Dialogue on AI Governance, UN Secretary-General António Guterres warned that AI is moving faster than the world’s institutions can respond.“An experiment is being run on our own societies — without a plan, and without consent,” warned Guterres, underlining that societies are being pushed into this new era without clear rules, shared safeguards, or meaningful public consent. The influence of AI across all spheres of human existence has skyrocketed over the past two years. It reshaped many workplaces, education, information consumption, and daily life. The whole industry is reviewing labour practices and hiring standards, with people’s productivity unfairly compared to AI's.Vulnerable communities are even more at risk of people using AI to spread hate and misogyny. “Abusers now can create and spread deepfakes, sexualized synthetic images, impersonation content, and other forms of image-based abuse faster, cheaper, and with less technical skill,” explained Kalliopi Mingeirou, UN Women.But experts speak not only of dangers but also of the world of possibilities that AI opens for people and societies, proposing access to knowledge, skills, and solutions they hadn’t before. AI can help doctors detect illness faster. It can support children who need extra help with learning. It can give farmers better forecasts. It can support sustainable recovery and development.“Used well, and shared widely, AI could compress decades of development into years,” said Secretary-General. “For countries rebuilding, modernizing, and defending democratic institutions, responsible AI could become a powerful tool.”“Innovations need guardrails”The message is simple: innovation needs rules. We trust aviation not because it couldn’t be dangerous, but because it has rules. Same with cars. Imagine not having any traffic regulations? A car is a tool, and rules are needed for those behind the wheel. Same with AI.“AI is a tool that millions of people can benefit from. But at the same time, if it’s in the bad hands, it could also be used for coercive purposes, to erode trust, undermine democratic structures, and it could be used for propaganda and against information integrity,” explained Rein Tammsaar, co-chair of the Global Dialogue on AI governance, Permanent Representative of Estonia to the United Nations. UN Secretary-General cautions that many nations have had little say in decisions that may shape their futures. Therefore, the UN is becoming a platform for Member States to discuss and set up priorities for global action based on safety, red lines, capacity, and transparency.Safety, especially for children, is paramount, while human rights are at the core in areas such as policing, healthcare, justice, and social services.“No child should be a guinea pig for unregulated AI.”Countries must not be excluded from the AI revolution because they lack computing power, data, skills, or investment. The UN plans to support networks and funding mechanisms to help countries build AI capacity, especially in developing countries. Across the UN system, AI is already being used to help people in practical ways: helping the World Food Programme deliver food faster and more efficiently, supporting UNICEF and ITU’s initiative in connecting young people to the internet. It also helps to spot risks, detect crises earlier, make networks safer, and guide teams working in difficult conditions.Ukrainian contextUkraine continues digital transformation, recovery, and European integration against the backdrop of the highly challenging circumstances of the large-scale invasion of Russia.Ukrainians live in an information environment where truth is constantly under attack. Disinformation, fake videos, manipulated images, and automated online campaigns are part of the wider reality of war. The usage of AI deepens these issues, while its application in military drones poses real dangers for the civilian population.However, used well, AI tools can help rebuild and protect society. The Ukrainian government has already made some progress in using AI for mine clearance, while further application of AI tools could support reconstruction, public services, education, healthcare, and democratic resilience.The future of AI is still open. Now the task is to build that future by humanity, with humanity, and for all humanity.Further information on AI Global Dialogue is available at the link.Speech of the UN Secretary-General during the opening session.