Paper by Harini Suresh et al: “Growing interest and investment in the capabilities of foundation models has positioned such systems to impact a wide array of services, from banking to healthcare. Alongside these opportunities is the risk that these systems reify existing power imbalances and cause disproportionate harm to historically marginalized groups. The larger scale and domain-agnostic manner in which these models operate further heightens the stakes: any errors or harms are liable to reoccur across use cases. In AI & ML more broadly, participatory approaches hold promise to lend agency and decision-making power to marginalized stakeholders, leading to systems that better benefit justice through equitable and distributed governance. But existing approaches in participatory AI/ML are typically grounded in a specific application and set of relevant stakeholders, and it is not straightforward how to apply these lessons to the context of foundation models. Our paper aims to fill this gap.
First, we examine existing attempts at incorporating participation into foundation models. We highlight the tension between participation and scale, demonstrating that it is intractable for impacted communities to meaningfully shape a foundation model that is intended to be universally applicable. In response, we develop a blueprint for participatory foundation models that identifies more
local, application-oriented opportunities for meaningful participation. In addition to the “foundation” layer, our framework proposes the “subfloor” layer, in which stakeholders develop shared technical infrastructure, norms and governance for a grounded domain such as clinical care, journalism, or finance, and the “surface” (or application) layer, in which affected communities shape the use of a foundation model for a specific downstream task. The intermediate “subfloor” layer scopes the range of potential harms to consider, and affords communities more concrete avenues for deliberation and intervention. At the same time, it avoids duplicative effort by scaling input across relevant use cases. Through three case studies in clinical care, financial services, and journalism, we illustrate how this multi-layer model can create more meaningful opportunities for participation than solely intervening at the foundation layer…(More)”.
The citizen’s panel on AI issues its report
Belgian presidency of the European Union: “Randomly select 60 citizens from all four corners of Belgium. Give them an exciting topic to explore. Add a few local players. Season with participation experts. Bake for three weekends at the Egmont Palace conference centre. And you’ll end up with the rich and ambitious views of citizens on the future of artificial intelligence (AI) in the European Union.
This is the recipe that has been in progress since February 2024, led by the Belgian presidency of the European Union, with the ambition of involving citizens in this strategic field and enriching the debate on AI, which has been particularly lively in recent months as part of the drafting of the AI Act recently adopted by the European Parliament.
And the initiative really cut the mustard, as the 60 citizens worked enthusiastically, overcoming their apprehensions about a subject as complex as AI. In a spirit of collective intelligence, they dove right into the subject, listening to speakers from academia, government, civil society and the private sector, and sharing their experiences and knowledge. Some of them were just discovering AI, while others were already using it. They turned this diversity into a richness, enabling them to write a report on citizens’ views that reflects the various aspirations of the Belgian population.
At the end of the three weekends, the citizens almost unanimously adopted a precise and ambitious report containing nine key messages focusing on the need for a responsible, ambitious and beneficial approach to AI, ensuring that it serves the interests of all and leaves no one behind…(More)”
Dynamic Collective Action and the Power of Large Numbers
Paper by Marco Battaglini & Thomas R. Palfrey: “Collective action is a dynamic process where individuals in a group assess over time the benefits and costs of participating toward the success of a collective goal. Early participation improves the expectation of success and thus stimulates the subsequent participation of other individuals who might otherwise be unwilling to engage. On the other hand, a slow start can depress expectations and lead to failure for the group. Individuals have an incentive to procrastinate, not only in the hope of free riding, but also in order to observe the flow of participation by others, which allows them to better gauge whether their own participation will be useful or simply wasted. How do these phenomena affect the probability of success for a group? As the size of the group increases, will a “power of large numbers” prevail producing successful outcomes, or will a “curse of large numbers” lead to failure? In this paper, we address these questions by studying a dynamic collective action problem in which n individuals can achieve a collective goal if a share of them takes a costly action (e.g., participate in a protest, join a picket line, or sign an environmental agreement). Individuals have privately known participation costs and decide over time if and when to participate. We characterize the equilibria of this game and show that under general conditions the eventual success of collective action is necessarily probabilistic. The process starts for sure, and hence there is always a positive probability of success; however, the process “gets stuck” with positive probability, in the sense that participation stops short of the goal. Equilibrium outcomes have a simple characterization in large populations: welfare converges to either full efficiency or zero as n→∞ depending on a precise condition on the rate at which the share required for success converges to zero. Whether success is achievable or not, delays are always irrelevant: in the limit, success is achieved either instantly or never…(More)”
Cities Are at the Forefront of AI and Civic Engagement
Article by Hollie Russon Gilman and Sarah Jacob: “…cities worldwide are already adopting AI for everyday governance needs. Buenos Aires is integrating communication with residents through Boti, an AI chatbot accessible via WhatsApp. Over 5 million residents are using the chatbot everyday month, with some months upwards of 11 million users. Boti connects residents with city services such as bike sharing or social care programs or reports. Unlike other AI systems with a closed loop, Boti can connect externally to help residents with other government services. For more sensitive issues, such as domestic abuse, Boti can connect residents with a human operator. AI, in this context, offers residents a convenient means to efficiently engage with city resources and communicate with city employees.
Another example of AI improving people’s everyday lives is SomosUna, a partnership between the Inter American Development Bank and Next2MyLife, aims to address gender-based violence in Uruguay. In response to the rise in gender-based violence during and after Covid, this initiative aims to prevent violence through a network of support and “helpers” which includes 1) training 2) technology and 3) a community of volunteers. This initiative will leverage AI technology to enhance its support network, advancing preventative measures and providing immediate assistance.
While AI can foster engagement, local government officials recognize that they must pre-engage the public to determine the role that AI should play in civic life across diverse cities. This pre-engagement and education will inform the ethical standards and considerations against which AI will be assessed.
The EU’s ITHACA project, for example, explores the application of AI in civic participation and local governance…(More)”… See also: AI Localism.
Democratic innovations beyond the deliberative paradigm
Paper by Christian Opitz: “The current research on deliberative-participatory democratic innovations conducted by state administration agencies exhibits empirical eclecticism and is dominated by a deliberative paradigm. However, this paradigm tends to conflate normative prescription with analytical description. In contrast, this article proposes a comprehensive re-conceptualization of such innovations, drawing from Niklas Luhmann’s systems theory. It outlines the specific problem these innovations address (function), how they operate in tackling this problem (functioning) and the problems they inevitably raise (dysfunctions). In addition, my re-conceptualization retains the possibility to critically compare these (and other) experiments regarding their capability to address emerging challenges within the modern democratic political system…(More)”.
Social Choice for AI Alignment: Dealing with Diverse Human Feedback
Paper by Vincent Conitzer, et al: “Foundation models such as GPT-4 are fine-tuned to avoid unsafe or otherwise problematic behavior, so that, for example, they refuse to comply with requests for help with committing crimes or with producing racist text. One approach to fine-tuning, called reinforcement learning from human feedback, learns from humans’ expressed preferences over multiple outputs. Another approach is constitutional AI, in which the input from humans is a list of high-level principles. But how do we deal with potentially diverging input from humans? How can we aggregate the input into consistent data about ”collective” preferences or otherwise use it to make collective choices about model behavior? In this paper, we argue that the field of social choice is well positioned to address these questions…(More)”.
The generation of public value through e-participation initiatives: A synthesis of the extant literature
Paper by Naci Karkin and Asunur Cezar: “The number of studies evaluating e-participation levels in e-government services has recently increased. These studies primarily examine stakeholders’ acceptance and adoption of e-government initiatives. However, it is equally important to understand whether and how value is generated through e-participation, regardless of whether the focus is on government efforts or user adoption/acceptance levels. There is a need in the literature for a synthesis focusing on e- participation’s connection with public value creation using a systematic and comprehensive approach. This study employs a systematic literature review to collect, examine, and synthesize prior findings, aiming to investigate public value creation through e-participation initiatives, including their facilitators and barriers. By reviewing sixty-four peer-reviewed studies indexed by Web of Science and Scopus, this research demonstrates that e-participation initiatives and efforts can generate public value. Nevertheless, several factors are pivotal for the success and sustainability of these initiatives. The study’s findings could guide researchers and practitioners in comprehending the determinants and barriers influencing the success and sustainability of e-participation initiatives in the public value creation process while highlighting potential future research opportunities in this domain…(More)”.
Designing Digital Voting Systems for Citizens
Paper by Joshua C. Yang et al: “Participatory Budgeting (PB) has evolved into a key democratic instrument for resource allocation in cities. Enabled by digital platforms, cities now have the opportunity to let citizens directly propose and vote on urban projects, using different voting input and aggregation rules. However, the choices cities make in terms of the rules of their PB have often not been informed by academic studies on voter behaviour and preferences. Therefore, this work presents the results of behavioural experiments where participants were asked to vote in a fictional PB setting. We identified approaches to designing PB voting that minimise cognitive load and enhance the perceived fairness and legitimacy of the digital process from the citizens’ perspective. In our study, participants preferred voting input formats that are more expressive (like rankings and distributing points) over simpler formats (like approval voting). Participants also indicated a desire for the budget to be fairly distributed across city districts and project categories. Participants found the Method of Equal Shares voting rule to be fairer than the conventional Greedy voting rule. These findings offer actionable insights for digital governance, contributing to the development of fairer and more transparent digital systems and collective decision-making processes for citizens…(More)”.
Citizen Jury on New Genomic Techniques
Paper by Kai P. Purnhagen and Alexandra Molitorisova: “Between 26-28 January 2024, a citizen jury was convened at the Schloss Thurnau in Upper Franconia, Germany to deliberate about new genomic techniques (NGTs) used in agriculture and food/feed production, ahead of the vote of the European Parliament and the Council of the European Union on the European Commission’s proposal for a regulation on plants obtained by certain NGTs and their food and feed. This report serves as a policy brief with all observations, assessments, and recommendations agreed by the jury with a minimum of 75 percent of the jurors’ votes. This report aims to provide policymakers, stakeholders, and the public with perspectives and considerations surrounding the use of NGTs in agriculture and food/feed production, as articulated by the members of the jury. There are 18 final recommendations produced by the jury. Through thoughtful analysis and dialogue, the jury sought to contribute to informed decision-making processes…(More)”.
Making Sense of Citizens’ Input through Artificial Intelligence: A Review of Methods for Computational Text Analysis to Support the Evaluation of Contributions in Public Participation
Paper by Julia Romberg and Tobias Escher: “Public sector institutions that consult citizens to inform decision-making face the challenge of evaluating the contributions made by citizens. This evaluation has important democratic implications but at the same time, consumes substantial human resources. However, until now the use of artificial intelligence such as computer-supported text analysis has remained an under-studied solution to this problem. We identify three generic tasks in the evaluation process that could benefit from natural language processing (NLP). Based on a systematic literature search in two databases on computational linguistics and digital government, we provide a detailed review of existing methods and their performance. While some promising approaches exist, for instance to group data thematically and to detect arguments and opinions, we show that there remain important challenges before these could offer any reliable support in practice. These include the quality of results, the applicability to non-English language corpuses and making algorithmic models available to practitioners through software. We discuss a number of avenues that future research should pursue that can ultimately lead to solutions for practice. The most promising of these bring in the expertise of human evaluators, for example through active learning approaches or interactive topic modeling…(More)”.