citiesabc, first_page
Governing AI for Cities: Why the UN Is Calling for Global Rules on Safety, Children, Climate and Human Control
13 Aug 2026
The United Nations is pushing for stronger international cooperation on artificial intelligence as AI becomes embedded in public services, urban infrastructure, education and everyday life.
Artificial intelligence is becoming part of the infrastructure of modern society.
Cities are introducing AI into transport, healthcare, public administration, education, security and environmental management. At the same time, data centres supporting AI systems are increasing their demands for energy, water and land.
Against this background, the United Nations is calling for a more coordinated approach to global AI governance.
At the inaugural UN Global Dialogue on Artificial Intelligence Governance in Geneva on 6 July 2026, Secretary-General António Guterres argued that AI is developing faster than institutions are adapting to its consequences. The Dialogue brings governments together with technology companies, academics, civil society and technical experts to establish common approaches to governing AI.
The debate matters for cities because many of the consequences of AI, from data-centre construction to automated public services, will ultimately be experienced locally.
Four Priorities for Responsible AI
Guterres outlined four major areas requiring international attention: safety, human rights, capacity and transparency.
The UN's wider argument is that AI governance cannot be determined solely by the countries and technology companies currently leading AI development.
The Global Dialogue was created as a UN platform where all Member States can participate, alongside other stakeholders. It builds on the Global Digital Compact and a General Assembly decision establishing new international AI-governance mechanisms.
A central concern is inequality.
The UN warns that AI could accelerate progress in areas including healthcare, education, agriculture, scientific research and disaster preparedness. But access to computing infrastructure, expertise and investment remains heavily concentrated, creating the risk that an existing digital divide becomes an even deeper AI divide.
AI's Environmental Footprint Is Becoming an Urban Issue
One of the most immediate concerns is the physical infrastructure behind artificial intelligence.
AI may feel like an invisible digital service, but running advanced systems requires large data centres, electricity generation, cooling systems and extensive hardware supply chains.
Research from the United Nations University estimates that data centres could consume around 945 terawatt-hours of electricity annually by 2030. AI-related water consumption could also reach the equivalent of the basic annual domestic water needs of around 1.3 billion people by the end of the decade.
These impacts are not evenly distributed.
Data centres are built in particular communities, where their electricity and water requirements interact with local grids, housing development, land use and water resources.
This makes AI infrastructure increasingly relevant to urban planning and environmental governance.
Guterres has consequently called for an AI Environmental Transparency Initiative, asking major AI companies to disclose the carbon, water and land footprints associated with their systems and to move data centres towards renewable energy by 2030.
For city governments, greater transparency could become important when deciding where data centres should be located and how their environmental costs should be managed.
Protecting Children in an AI-Enabled City
AI governance is also becoming a question of child safety.
Children increasingly encounter AI through education platforms, social networks, conversational assistants and entertainment services.
At the Global Dialogue, Guterres called for an AI Child Safety Pledge centred around three principles: companies should conduct child-specific safety testing before deployment, prevent systems from generating sexual-abuse material involving children, and ensure that children showing signs of crisis can be connected to human support.
For cities and education systems, this could influence how AI tools are introduced into schools, libraries and youth services.
The issue extends beyond whether children should be allowed to use AI.
The larger question is what responsibilities governments, technology providers and educational institutions have when AI systems become part of children's everyday environments.
Some Decisions Must Remain Human
The UN is also drawing a clearer boundary around autonomous weapons.
Guterres has reiterated his call for lethal autonomous weapons systems operating without meaningful human control to be prohibited under international law.
His argument is based on a simple principle: decisions involving taking human life should remain under human judgment and accountability.
Although this debate primarily concerns military systems, the underlying principle has wider relevance for smart cities.
AI systems are increasingly used to recommend or influence decisions in policing, welfare, employment, healthcare and public administration.
The more significant a decision becomes, the more important questions of human oversight, accountability and appeal become.
A smart city cannot simply be defined by how many decisions it automates.
It must also decide which decisions should never be delegated completely to machines.
The AI Divide Could Become a City Divide
Another major concern is access.
UN research shows that AI infrastructure, investment and expertise remain concentrated in relatively few countries. More than 90% of specialised AI computing capacity is estimated to be located in the United States and China, while many countries have little significant domestic AI infrastructure.
The same pattern could emerge within countries.
Cities containing leading universities, technology companies, data centres and highly skilled workforces may capture a disproportionate share of the benefits from AI.
Other regions could become dependent on technologies developed elsewhere without having the skills, infrastructure or governance capacity to shape how those systems affect their communities.
The UN is therefore proposing initiatives including a Global Fund for AI and a global network for AI capacity building aimed particularly at developing countries.
For citiesabc, this raises a larger question:
Could AI capability become one of the defining inequalities between cities in the next decade?
From Smart Cities to Responsible AI Cities
For years, smart-city strategies have focused on connectivity, sensors, data platforms and automation.
The next phase may require a different measure of technological progress.
A genuinely AI-ready city may need:
- transparent AI procurement;
- human oversight of high-impact decisions;
- AI literacy for public officials and citizens;
- environmental standards for digital infrastructure;
- strong protection for children and vulnerable groups;
- accessible mechanisms for challenging automated decisions;
- and public participation in how AI systems are introduced.
- Technology alone does not make a city intelligent.
- Governance determines whether technology strengthens public life or simply introduces new forms of risk.
- The UN's emerging AI-governance framework reflects that shift.
The central challenge is no longer simply whether artificial intelligence can make cities more efficient.
It is whether cities, governments and technology companies can ensure AI remains accountable to the people and communities it increasingly affects.
Sources
- United Nations — Secretary-General's Remarks at the First Global Dialogue on Artificial Intelligence Governance
https://press.un.org/en/2026/sgsm23211.doc.htm - United Nations — Global Dialogue on Artificial Intelligence Governance
https://www.un.org/global-dialogue-ai-governance/ - United Nations — Global Dialogue Press Release, 6 July 2026
https://www.un.org/global-dialogue-ai-governance/sites/default/files/2026-07/global_dialogue_press_release_en_0.pdf - United Nations — AI's Environmental Costs Threaten Water, Land and Climate
https://india.un.org/en/316754-ai%E2%80%99s-environmental-costs-threaten-water-land-and-climate - United Nations University — Environmental Cost of AI's Energy Use: Carbon, Water and Land Footprints
https://digitallibrary.un.org/record/4116063 - United Nations — Climate Crisis: UN Chief Calls for Transparency on AI Data Centres
https://www.un.org/en/desa/climate-crisis-un-chief-lays-out-solutions-blueprint-for-clean-energy-transition - United Nations — Independent International Scientific Panel on Artificial Intelligence
https://www.un.org/independent-international-scientific-panel-ai/ - UNCTAD — Technology and Innovation Report 2025
https://unctad.org/publication/technology-and-innovation-report-2025
