Artificial intelligence (AI) is quickly becoming part of everyday urban life. It is changing how residents access services, how children learn, how civil servants work and how cities respond to complex challenges, from mobility and climate adaptation to social inclusion and public safety.
For local governments, the question is no longer whether AI will be used, but how it can be used responsibly, inclusively and in ways that improve people’s lives.
Across Europe, cities are moving from experimentation to implementation. They are testing AI assistants, building data platforms, developing ethical frameworks, training public servants and exploring how AI can strengthen public services. But their experience points to a clear lesson: technology alone is not enough.
AI can only support better services if it is built on trust, good data, strong governance, public-sector capacity and a clear understanding of residents’ needs.
From tools to public value
In the city of Espoo, AI is being tested as a way to improve internal services and support municipal staff. The city is piloting an intranet AI assistant to help its 12,000 employees find internal information more quickly, combining GPT-based solutions and Microsoft Copilot to improve the efficiency and usability of internal information.
For Espoo, the key lesson is that AI must be tested carefully before being scaled. “Data quality and structure matter and experts and end users must be involved early,” states Anni Orttenvuori-Ganter, Lead Data and Knowledge Architect at the City of Espoo.
The city is also exploring emerging technologies in education and recruitment, including Minecraft-based sustainability learning and virtual reality for staff orientation. But for Orttenvuori-Ganter, the purpose matters more than the tool. “What’s important here is not the technology itself, but how it is used and for what purpose,” she says.
What’s important here is not the technology itself, but how it is used and for what purpose.
This is a message shared by many cities: AI should not be introduced because it is the latest trend, but because it helps solve a real problem. The starting point must be public value.
Trust, ethics and human oversight
As AI becomes more embedded in public administration, cities are facing harder questions about ethics, bias, accountability and control. The Hague is taking a strategy-led approach, using AI to improve services while managing risks.
“Our AI strategy helps make sure that what we do is done correctly, ethically and according to the right policy guidelines,” explains Michael Stam, Senior Policy Advisor European Affairs at the City of The Hague.
In practice, this means making deliberate choices about where AI should and should not be used. The Hague rejected the use of AI to classify job applicants because the risk of bias was too high, while continuing to explore AI in areas such as service delivery.
In the city of Riga, AI is being used in traffic control, where systems can help prepare evidence while the final decision on penalties remains with a police inspector. This reflects a wider principle for cities: AI can support decision-making, but public accountability must remain with people.
AI can help cities work faster and make better use of information, but it cannot replace public responsibility.
“AI can help cities work faster and make better use of information, but it cannot replace public responsibility,” says Arnis Gulbis, Director of Riga Digital Agency. “For Riga, the priority is to use AI in ways that are secure, practical and supported by human oversight.”
Data as the foundation
For cities, the quality of AI depends on the quality of data. Without reliable, interoperable and well-governed data, AI tools risk producing poor, biased or misleading results.
Cities are increasingly treating data as critical urban infrastructure. Riga’s work on mobility data, sensors and digital twins shows how cities are connecting information across departments to support better planning and decision-making.
However, this work brings technical and organisational challenges. Data must be stored in consistent ways, exchanged between systems and made usable across departments and levels of government.
European data spaces could play a major role in addressing these challenges. By connecting local data ecosystems across sectors such as mobility, energy, climate and urban planning, they can help cities achieve interoperability, improve collaboration and make innovation more sustainable.
“Data spaces are not only a technical infrastructure,” says Sophie Meszaros, Interoperability Strategy Lead at Open and Agile Smart Cities and Communities (OASC). “They are a way to help cities collaborate across sectors, connect local innovation with European priorities and make data usable in ways that support real services.”
For the city of Oulu, data spaces are also linked to trust and Europe’s wider approach to AI. The city sees the future in distributed and federated models, where data owners retain control.
“Trust is the basis for a European approach to AI,” says Ari Saine, Smart City Manager at Business Oulu. “Without clear rules, there is no trust. And without trust, cities cannot build the data spaces and services that allow AI to create real public value.”
Without clear rules, there is no trust. And without trust, cities cannot build the data spaces and services that allow AI to create real public value.
Digital sovereignty and practical tools
As cities adopt AI, they are also asking who controls the systems they rely on.
Digital sovereignty – who has control over digital tools and systems – has become a growing concern, particularly as local governments depend heavily on large global technology providers. This affects procurement, cybersecurity, data protection, cloud infrastructure, staff skills and democratic control.
“Cities need AI solutions they can trust, understand and govern,” says Pieter-Jan Pauwels, AI Program Manager and International Policy Advisor at District09, Ghent. “That means looking not only at what AI can do, but also at who controls the tools, how data is protected and how public administrations remain accountable.”
Cities are also clear that strategy alone is not enough. They need practical tools, shared examples and peer support to move from principles to implementation in the full respect of complex EU legislation. Priorities include secure solutions, impact measurement, use cases and clearer thinking about when AI is genuinely needed.
The message from city governments is clear: AI is powerful, but it is not needed for everything. Cities need to be focused on the problem they are trying to solve, the risks involved and the public value they want to create.
Skills, culture and inclusion
Responsible AI also depends on people. Cities need public servants who understand how to use AI, managers who can govern it, technical teams who can assess risks, and residents who can take part in an increasingly digital society.
In Bulgaria, the Institute of Public Administration has developed an AI Readiness Index to assess how prepared public institutions are to adopt AI. The average score for Bulgarian public administrations taking the assessment was 49 out of 100, showing that institutions are still building capacity, with common gaps being identified around leadership engagement, internal AI policies and awareness of the AI Act.
At the same time, many public servants are eager to learn how AI can be applied. “It is not if, but how we are adopting and applying AI throughout the public sector,” says Tanya Ivanova-Chikova, Training Manager at Institute of Public Administration.
But AI readiness cannot be reduced to technical training alone. Bordeaux Metropole is focusing not only on digital skills, but on building a broader digital culture.
Skills are about using digital tools. Culture is about understanding what those tools are, how they shape our lives and what choices we have.”
“Digital culture is not the same as digital skills,” says Sophie Woodville, Digital Program Manager at Bordeaux Metropole and Technical Chair of the Eurocities Digital Forum. “Skills are about using digital tools. Culture is about understanding what those tools are, how they shape our lives and what choices we have.”
For Bordeaux Metropole, this means helping residents ask critical questions about technology, data and alternatives, so that digital transformation is something they can understand, debate and help shape.
Supporting residents and civil society
Cities are also helping residents and local organisations navigate an AI-shaped world.
In Linkoping, AI is becoming part of the education system, with teachers using it to create learning materials, quizzes and interactive assignments. But the city’s focus is not only on using tools.
In Amsterdam, work on digital citizenship includes protecting young people online and understanding how online harms affect life offline. Pedro Campos Ponce, Coordinator Team EU & International at Amsterdam, says cities are seeing “offline consequences of the online world”, including impacts on mental health, cyberbullying and conflicts that begin online.
In Manchester, digital citizenship is linked to inclusion, infrastructure and representation, with a focus on digital skills, digital equity and connecting communities.
Leipzig, meanwhile, is supporting NGOs, clubs and associations that often work directly with residents but lack digital capacity themselves.
Looking ahead
The rapid development of AI is pushing cities to think beyond today’s tools. AI systems are beginning to move from analysis towards decision-making and, in some cases, autonomous action. New developments, from AI agents to quantum technologies, could reshape assumptions about security, data and computational power.
The future digital city will depend on whether cities can organise data, trust and skills in ways that keep people and public value at the centre.
“The future digital city will not be defined by AI alone,” says Jochem Cooiman, Innovation Officer at the City of Rotterdam. “It will depend on whether cities can organise data, trust and skills in ways that keep people and public value at the centre.”
AI offers major opportunities for cities. It can improve services, reduce administrative burdens, support learning, strengthen civil society and help local governments respond to complex challenges.
However, it also brings real risks. Without governance, trust, skills and democratic oversight, AI could reinforce inequalities, reduce transparency and deepen dependency on systems cities do not control.
The lesson from Europe’s cities is that AI should not be treated as a shortcut to transformation. It should be shaped as a public tool, guided by local needs, democratic values and the common good.
_____________
These discussions took place at the 2026 edition of the Eurocities Digital Forum: ‘Fostering a People Centred Digital Tomorrow,’ hosted by Sofia Municipality, on 15-17 April 2026. This annual event provides cities with the opportunity to share best practices, strengthen their advocacy towards the EU institutions, and explore innovative digital governance models.
Read Eurocities’ article on the Forum’s opening session: Building a common digital future for cities.
And our article on the forum discussions: Connecting data and services to build people-centred digital cities
All the photos from the forum can be found here.
