Artificial Intelligence and the Future of Citizen Participation

1. Introduction: Citizen participation in the age of artificial intelligence

Abstract

Citizen participation is a strategic asset for governments to design and deliver better policies while reinforcing citizens’ trust. Governments around the world have been leveraging digital technologies to improve and expand their capacity to design and implement meaningful citizen participation processes. Artificial intelligence constitutes a new frontier of civic technologies. This chapter provides an introduction to the role of digital technologies in citizen participation, with a focus on the potential benefits related to the adoption of AI systems.

Citizen participation for more effective policies and greater public trust

By providing citizens with meaningful and impactful opportunities to participate in policy and decision-making, governments at all levels can strengthen public trust. The global challenges and disruptions faced by OECD member and partner countries in recent years have eroded citizens’ trust in government. Only 39% of people report high or moderately high levels of trust, with an overall decline of 2 percentage points since 2020 (OECD, 2022[1]; OECD, 2024[2]). However, the feeling of political agency, defined as a combination of “feeling confident to participate in politics” and “feeling confident of being heard”, is strongly correlated with significantly higher levels of trust in governments. 69% of those who feel they have a say in government decisions report high trust in governments, as opposed to only 22% among those who do not. The gap in trust in government associated with the political agency (or the lack thereof) is wider than those associated with other social factors, such as socio-economic characteristics, education, and age. In other words, creating opportunities to reinforce political agency of citizens bears the potential of significantly improving trust in government.

Although decisions ultimately remain with elected representatives, governments can involve citizens in policymaking beyond and between elections to better understand citizens’ needs, and design and implement more effective policies, enhance the legitimacy and the acceptability of their decisions, and ultimately reinforce citizens’ trust (OECD, 2022[3]; Bertelsmann Stiftung, 2018[4]). Through the OECD (2017[5]) Recommendation on Open Government, all OECD member countries and several non-members have committed to promote such engagement – defined as “all the ways in which stakeholders (including citizens) can be involved in the policy cycle and in service design and delivery” - to design and deliver better policies while reinforcing citizens’ trust (OECD, 2023[6]). This definition of citizen participation comprises a ladder of three progressive levels (OECD, 2001[7]): information, consultation, and engagement (OECD, 2017[5]). Citizen participation processes include a variety of methods and mechanisms, ranging from access to information, to public consultations, to participatory budgeting to representative deliberative processes. Table 1.1 provides definitions and examples for each level of the ladder of participation. Governments across OECD member countries and beyond are creating a growing number of opportunities for citizens to have a say in policymaking. The OECD has collected more than 800 cases of representative deliberative processes since 1971 (OECD, 2023[8]), while the global network Participedia has compiled 2 272 cases of citizen participation processes since 2009 (Participedia, 2025[9]).

Table 1.1. Levels of citizen participation in policymaking

Level	Definition	Examples of citizen participation processes
Information	An initial level of participation characterized by a one-way relationship in which the government produces and delivers information to citizens and stakeholders. It covers both on-demand provision of information and “proactive” measures by the government to disseminate information.	Access to information Disclosure of public data
Consultation	A more advanced level of participation that entails a two-way relationship in which citizens and stakeholders provide feedback to the government and vice-versa. It is based on the prior definition of the issue for which views are being sought and requires the provision of relevant information, in addition to feedback on the outcomes of the process.	Townhall meetings Public consultations
Engagement	When citizens and stakeholders are given the opportunity and the necessary resources (e.g., information, data, and digital tools) to collaborate during all phases of the policy-cycle and in the service design and delivery. It acknowledges equal standing for citizens in setting the agenda, proposing project or policy options and shaping the dialogue – although the responsibility for the final decision or policy formulation in many cases rests with public authorities.	Citizen Science Civic monitoring Open innovation Participatory Budgeting Representative Deliberative Processes.

Note: the examples of citizen participation processes provided in this table are not exhaustive.

Source: (OECD, 2022[3]).

Digital technologies for citizen participation

Digital technologies offer significant opportunities to enhance and mainstream citizen participation in policymaking, namely by expanding the reach of participatory opportunities and complementing existing in-person mechanisms with new forms and channels of interactions between governments and citizens (OECD, 2022[3]). The OECD designates this field as Civic Tech, and defines it as “the use of digital technologies to reinforce democracy by enabling the public to be informed, participate in decision and policymaking, and increase governments’ responsiveness and accountability” (OECD, forthcoming[10]; 2025[11]). Digital tools are regularly used by public authorities at all levels of government to collect and analyse citizens’ contributions in consultations, to inform them about participation opportunities and in-person events, to provide learning materials, to perform online voting and to follow-up on the implementation of the results of participatory and deliberative processes (García and Al., 2023[12]; OECD, 2023[6]; People Powered, 2025[13]). Among the 43 Adherents to the OECD Recommendation on Open Government, 31 have developed a government-wide participation portal to centralise information on past, present, and future participation opportunities (OECD, 2023[6]). For example, Portugal developed Participa.gov, its one-stop-shop platform to enable citizen to take part in all government-led participatory processes. Beyond central governments, regions and cities across OECD member countries are at the forefront of democratic innovations, including through the deployment and at times the development of their own digital platforms for citizen participation. Since 2011, the city of Reykjavik, Iceland has been using the platform Better Reykjavik based on the open-source software Your Priorities (OECD Observatory on Public Sector Innovation, 2010[14]), while in 2016 Madrid and Barcelona in Spain deployed their participatory platforms Decide Madrid and Decidim Barcelona (OECD, 2022[3]), respectively based on the open-source software Consul and Decidim. Governments are also using digital tools for participation to complement processes that take place primarily in person. Box 1.1 showcases the experience of the Forum Against Fakes, a hybrid participatory process, organised in Germany in 2024 by the Bertelsmann Stiftung,

Box 1.1. The Forum Against Fakes in Germany

In 2024, the German Foundation Bertelsmann Stiftung designed and implemented the Forum Against Fakes in collaboration with the Federal Ministry of the Interior of Germany, to involve citizen in addressing the consequences of intentionally spread false or misleading information in society. This hybrid participatory process included a deliberative assembly of 120 randomly selected citizens from all parts of Germany and a digital platform designed by Make.org. Three online consultations were conducted in parallel of the work of the deliberative assembly to: (1) collect themes and proposals in relation to the question “Fakes and the manipulation of information: what should we do to protect ourselves and our democracy?” before the beginning of the in-person citizens’ assembly, (2) provide feedback on provisional recommendations proposed by the members of the assembly, and (3) vote on the final recommendations. In total, nearly 424,000 citizens participated, submitting more than 3,300 proposals and comments and casting more than 1,5 million votes. The process generated 28 concrete measures to implement 15 recommendations, which were presented to the Federal Ministry of the Interior.

Source: (Bertelsmann Stiftung, 2024[15]).

Governments are also exploring relevant uses of AI and emerging technologies to improve and enhance citizen participation. The OECD defines emerging technologies as those “characterised by rapid development, evolution, novelty and uncertainty in trajectory and impact” (OECD, 2024[16]). AI and emerging technologies such as blockchain and augmented or virtual reality (AR-VR) are among the technologies that generate opportunities to address citizen participation challenges (OECD, 2025[17]). For example, the city of Eindhoven in the Netherlands partnered with Eindhoven University of Technology to use Virtual Reality to engage with citizens in urban planning (Dane et al., 2024[18]), while Bellpuig in Spain, a municipality of about 5,000 inhabitants, conducted a local referendum allowing citizens to safely vote online through the blockchain based tool Vochain (Vocdoni, 2022[19]). While the uptake of technologies such as blockchain and AR-VR has remained limited (OECD, 2025[17]), the use of AI tools for citizen participation is a growing trend.

The potential of AI tools for participation

The OECD has adopted the following definition for AI: “an AI system is a machine-based system that, for explicit or implicit objectives, infers, from the input it receives, how to generate outputs such as predictions, content, recommendations, or decisions that can influence physical or virtual environments. Different AI systems vary in their levels of autonomy and adaptiveness after deployment.”1 Box 1.2 provides more details on the different types of AI systems in a historical perspective.

Box 1.2. The evolution of AI

Early AI systems

The evolution of AI began in the 1950s with Alan Turing's foundational question about machine intelligence. Early AI systems were rules-based, relying on programmed "If-then" logic (if a condition, then an action) to simulate intelligent behaviour. These systems required extensive human input and are still used today, such as in robotic process automation (RPA), though their limitations have led some to question whether they still qualify as true AI.

Machine Learning and Deep Learning

In the 21st century, breakthroughs in the branch of AI called machine learning (ML) changed the AI landscape by enabling systems to learn from data rather than explicit programming. ML uses algorithms and statistical models to identify patterns and improve performance over time through a process called "training". The combination of ML techniques, large datasets, and powerful computing has significantly expanded AI’s capabilities. Inspired by the human brain, neural networks are made up of layers of “neurons”, known as “nodes”, that process inputs with weights and biases to give specific outputs. A subset of algorithms in the area of neural networks - called deep neural networks (in the field of study and set of techniques called deep learning) - allows machine-based systems to “learn” from examples to make predictions or “inferences” based on the large amount of data processed during their training phase. The complexity of these models can make it difficult to understand their outputs and the processes leading to those.

Generative AI and Large Language Models

In 2017, Google researchers introduced a type of neural network architecture called “transformers”, which learn to detect how data elements - such as the words in this sentence - influence and depend on each other. This led to the rise of generative AI, including large language models (LLMs), such as those that power OpenAI’s ChatGPT (Chat Generative Pre-Trained Transformer), and the development of foundation models that can be adapted for various tasks. Although transformers are often discussed, other approaches exist, especially for non-text (e.g. images, video, audio) generation, such as generative adversarial network (GANs) and diffusion models.

Emerging forms of “agentic” AI

While most AI remains “narrow” (task-focused), some experts argue that foundation models are an early form of more “general” AI that can handle a broader array of tasks. AI experts and researchers are also increasingly exploring the potential of autonomous "agentic" systems that can handle entire flows of actions with little to no human supervision. Despite their potential, these systems are still early-stage and pose specific limitations and risks. As AI grows more capable, many experts advocate for human-AI collaboration rather than full reliance on machines.

Note: The OECD primer Hello World: Artificial intelligence and its use in the public sector (Berryhill et al., 2019[20]) provides details on the technical underpinnings and potential implications of AI.

Source: (OECD, 2025[11]; Bertelsmann Stiftung, 2025[21]).

Since 2019, the OECD has been studying the use of AI in government and the broader implications of this emerging trend. This has included foundational studies of the technology behind AI and its implications by and for governments (Berryhill et al., 2019[20]; Ubaldi, 2019[22]); targeted analyses of specific countries (OECD/CAF, 2022[23]); (OECD, 2024[24]; OECD/UNESCO, 2024[25]); and the assessment of global trends in government innovation, which often involve AI; and establishing a Framework for Trustworthy AI in Government (OECD, 2025[11]). OECD efforts on this topic can be found on a dedicated hub for AI in Government on the OECD.AI Policy Observatory at OECD.AI/gov.

Overall, the OECD has found that governments across OECD member countries and beyond are increasingly experimenting and mainstreaming the use of AI to improve their productivity and deliver more tailored and more timely public services (OECD, 2024[26]; OECD, 2023[27]). More specifically, the use of AI in the broader public sector responds to three priority objectives (OECD, 2019[28]):

• Productivity with more efficient internal operations and more effective policy design, decision-making, and service delivery. For example, using predictive AI systems for more effective policy planning, automating processes for more accelerated service delivery, and boosting performance by allowing civil servants to focus less on mundane tasks and more on mission-critical activities.

• Responsiveness of public policies and service through enhanced design and delivery approaches that better meet evolving needs of citizens and specific communities, and through improved civic participation mechanisms. This includes offering more and more proactively personalised, citizen-centric public services.

• Accountability by enhancing capacity for oversight and transparency. This shift may boost overall public satisfaction and enhance the perception of government as competent, fair, and responsive, ultimately strengthening public trust in government’s capacity for innovation and transformation.

The OECD report Governing with Artificial Intelligence: The State of Play and Way Forward in Core Government Functions builds on extensive research and a comprehensive analysis of use cases collected across 11 government functions. The results of the analysis emphasise a strong interest among governments across OECD member and partner countries to implement AI solutions for citizen participation and open government. Twenty-nine (14.5%) of the 200 use cases analysed in the report relate to citizen participation and open government, making it most represented government function after public service design and delivery (45 cases, 22.5% of the total). The public-facing nature of public services and citizen participation processes, their horizontality across government agencies and line ministries, and the public nature of the information shared may explain why public services and citizen participation are the most represented government functions in the emerging landscape of AI adoption in the public sector (OECD, 2025[11]).

To support governments in harnessing the benefits of AI while mitigating risks and overcoming implementation challenges, the report outlines the OECD Framework for Trustworthy AI in Government (see Figure 4.1). The framework defines Engagement of citizens and stakeholders as one of the key pillars to ensure that the adoption of AI by governments reflects the needs of all, showing that the intersection of AI and citizen participation should be seen as a two-way street. This resonates with the OECD Framework for Anticipatory Governance of Emerging Technologies, which stresses the importance of including ethical and value-based considerations, forward-looking approaches in assessing governance needs, meaningful societal engagement, agile regulation and international co-operation when bringing a new technology into society (OECD, 2024[16]). Although the OECD recognises AI as an “emerged” technology, these principles remain relevant to make sure that AI is developed and deployed with purpose.

This report deepens the research and analysis of the opportunities that AI tools offer to support governments and citizens when designing, implementing, communicating, and participating in citizen participation processes. It acknowledges the existing challenges of citizen participation, such as a disconnection from decision-making, as well as issues of accessibility, and scale. It also identifies where AI tools could help solve these challenges (OECD, 2025[17]). If implemented adequately, AI tools could provide citizens with more and better opportunities to participate in policymaking, enabling governments to deliver better policies and strengthen public trust (People Powered, 2025[13]). This report adopts the perspective of both governments and citizens as users of AI tools.

More specifically, it proposes a typology of applications of AI tools in support of tasks and activities that civil servants and citizens perform in the context of citizen participation. The typology, presented in Section 2, draws from desk research and from the conclusions of 50 use cases from 22 countries. Beyond task-specific AI tools, the typology provides a framework for discussing the future of AI systems applied to citizen participation in policymaking, with a focus on AI agents and “augmented collective intelligence”. Its purpose is to help governments and practitioners of participation navigate the opportunities offered by AI tools to address the challenges of citizen participation. It also acknowledges that the technology alone is not sufficient to make participation more meaningful and impactful, refusing to adopt a techno-solutionist approach.

References

[20] Berryhill, J. et al. (2019), “Hello, World: Artificial intelligence and its use in the public sector”, OECD Working Papers on Public Governance, No. 36, OECD Publishing, Paris, https://doi.org/10.1787/726fd39d-en.

[21] Bertelsmann Stiftung (2025), Public AI Whitepaper, https://doi.org/10.11586/2025040.

[15] Bertelsmann Stiftung (2024), Forum Against Fakes: Citizens’ report on how to deal with disinformation, https://doi.org/10.11586/2024149.

[4] Bertelsmann Stiftung (2018), Mitreden, Mitgestalten, Mitentscheiden: Five ideas for renewing democratic participation, https://www.bertelsmann-stiftung.de/de/publikationen/publikation/did/mitreden-mitgestalten-mitentscheiden-1.

[18] Dane, G. et al. (2024), “valuating a new framework for the participatory co-design of healthy public spaces using immersive virtual reality”, Computers, Environment and Urban Systems Volume 114, https://doi.org/10.1016/j.compenvurbsys.2024.102194.

[12] García, D. and E. Al. (2023), Democracy Technologies in Europe: Online Participation, Deliberation and Voting, Innovation in Politics Institute, https://www.idea.int/publications/catalogue/democracy-technologies-europe.

[11] OECD (2025), Governing with Artificial Intelligence: The state of play and way forward in core government functions, OECD Publishing.

[17] OECD (2025), “Tackling civic participation challenges with emerging technologies: Beyond the hype”, OECD Public Governance Policy Papers, No. 72, OECD Publishing, Paris, https://doi.org/10.1787/ec2ca9a2-en.

[29] OECD (2024), “Exploring New Frontiers of Citizen Participation in the Policy Cycle”, https://www.oecd.org/content/dam/oecd/en/about/programmes/reinforcing-democracy-initiative/Exploring-New-Frontiers-of-Citizen-Participation-Discussion-Paper.pdf.

[16] OECD (2024), “Framework for Anticipatory Governance of Emerging Technologies”, OECD Science, Technology and Industry Policy Papers, No. 165, OECD Publishing, Paris, https://doi.org/10.1787/0248ead5-en.

[26] OECD (2024), “Governing with Artificial Intelligence: Are governments ready?”, OECD Artificial Intelligence Papers, No. 20, OECD Publishing, Paris, https://doi.org/10.1787/26324bc2-en.

[24] OECD (2024), OECD Artificial Intelligence Review of Germany, https://doi.org/10.1787/609808d6-en.

[2] OECD (2024), OECD Survey on Drivers of Trust in Public Institutions – 2024 Results: Building Trust in a Complex Policy Environment, OECD Publishing, Paris, https://doi.org/10.1787/9a20554b-en.

[27] OECD (2023), Global Trends in Government Innovation 2023, OECD Public Governance Reviews, OECD Publishing, Paris, https://doi.org/10.1787/0655b570-en.

[8] OECD (2023), OECD Database of Representative Deliberative Processes, https://airtable.com/appP4czQlAU1My2M3/shrX048tmQLl8yzdc/tblrttW98WGpdnX3Y/viwX5ZutDDGdDMEep.

[6] OECD (2023), Open Government for Stronger Democracies: A Global Assessment, OECD Publishing, https://doi.org/10.1787/5478db5b-en.

[1] OECD (2022), Building Trust to Reinforce Democracy: Main Findings from the 2021 OECD Survey on Drivers of Trust in Public Institutions, Building Trust in Public Institutions, OECD Publishing, Paris, https://doi.org/10.1787/b407f99c-en.

[3] OECD (2022), OECD Guidelines for Citizen Participation Processes, OECD Public Governance Reviews, OECD Publishing, Paris, https://doi.org/10.1787/f765caf6-en.

[30] OECD (2020), “Innovative Citizen Participation and New Democratic Institutions: Catching the Deliberative Wave”, OECD Publishing, https://www.oecd.org/en/publications/innovative-citizen-participation-and-new-democratic-institutions_339306da-en.html.

[28] OECD (2019), Open Government Data.

[5] OECD (2017), Recommendation of the Council on Open Government.

[7] OECD (2001), Citizens as Partners: Information, Consultation and Public Participation in Policy-Making, OECD Publishing, Paris, https://doi.org/10.1787/9789264195561-en.

[10] OECD (forthcoming), Civic Tech for Democracy.

[14] OECD Observatory on Public Sector Innovation (2010), , https://oecd-opsi.org/innovations/better-reykjavik/.

[23] OECD/CAF (2022), The Strategic and Responsible Use of Artificial Intelligence in the Public, OECD Publishing, https://doi.org/10.1787/1f334543-en.

[25] OECD/UNESCO (2024), G7 Toolkit for Artificial Intelligence in the Public Sector, https://doi.org/10.1787/421c1244-en.

[9] Participedia (2025), Participedia, https://participedia.net/.

[13] People Powered (2025), Guide to Digital Participation Platform, https://drive.google.com/file/d/1J0BlXA5d6Fcgbk7MIZTqaq36eVgl9jaP/view?pli=1.

[22] Ubaldi, B. (2019), State of the art in the use of emerging technologies in the public sector, OECD Publishing, https://doi.org/10.1787/932780bc-en.

[19] Vocdoni (2022), First Public Institution in Spain Holds Vocdoni Referendum | Bellpuig Council, https://blog.vocdoni.io/referendum-bellpuig.

Note

← 1.  The OECD published further guidance on the definition of AI which can help the reader understand AI systems and its components: https://oecd.ai/en/wonk/definition.