Artificial intelligence (AI) systems are playing an overarching role in the disinformation phenomenon our world is currently facing. Such systems boost the problem not only by increasing opportunities to create realistic AI-generated fake content, but also, and essentially, by facilitating the dissemination of disinformation to a targeted audience and at scale by malicious stakeholders. This situation entails multiple ethical and human rights concerns, in particular regarding human dignity, autonomy, democracy, and peace. In reaction, other AI systems are developed to detect and moderate disinformation online. Such systems do not escape from ethical and human rights concerns either, especially regarding freedom of expression and information. Having originally started with ascending co-regulation, the European Union (EU) is now heading toward descending co-regulation of the phenomenon. In particular, the Digital Services Act proposal provides for transparency obligations and external audit for very large online platforms’ recommender systems and content moderation. While with this proposal, the Commission focusses on the regulation of content considered as problematic, the EU Parliament and the EU Council call for enhancing access to trustworthy content. In light of our study, we stress that the disinformation problem is mainly caused by the business model of the web that is based on advertising revenues, and that adapting this model would reduce the problem considerably. We also observe that while AI systems are inappropriate to moderate disinformation content online, and even to detect such content, they may be more appropriate to counter the manipulation of the digital ecosystem.

A lack of political legitimacy undermines the ability of the European Union (EU) to resolve major crises and threatens the stability of the system as a whole. By integrating digital data into political processes, the EU seeks to base decision-making increasingly on sound empirical evidence. In particular, artificial intelligence (AI) systems have the potential to increase political legitimacy by identifying pressing societal issues, forecasting potential policy outcomes, and evaluating policy effectiveness. This paper investigates how citizens’ perceptions of EU input, throughput, and output legitimacy are influenced by three distinct decision-making arrangements: (a) independent human decision-making by EU politicians; (b) independent algorithmic decision-making (ADM) by AI-based systems; and (c) hybrid decision-making (HyDM) by EU politicians and AI-based systems together. The results of a preregistered online experiment (n = 572) suggest that existing EU decision-making arrangements are still perceived as the most participatory and accessible for citizens (input legitimacy). However, regarding the decision-making process itself (throughput legitimacy) and its policy outcomes (output legitimacy), no difference was observed between the status quo and HyDM. Respondents tend to perceive ADM systems as the sole decision-maker to be illegitimate. The paper discusses the implications of these findings for (a) EU legitimacy and (b) data-driven policy-making and outlines (c) avenues for future research.

This article discusses the technology of city digital twins (CDTs) and its potential applications in the policymaking context. The article analyzes the history of the development of the concept of digital twins and how it is now being adopted on a city-scale. One of the most advanced projects in the field—Virtual Singapore—is discussed in detail to determine the scope of its potential domains of application and highlight challenges associated with it. Concerns related to data privacy, availability, and its applicability for predictive simulations are analyzed, and potential usage of synthetic data is proposed as a way to address these challenges. The authors argue that despite the abundance of urban data, the historical data are not always applicable for predictions about the events for which there does not exist any data, as well as discuss the potential privacy challenges of the usage of micro-level individual mobility data in CDTs. A task-based approach to urban mobility data generation is proposed in the last section of the article. This approach suggests that city authorities can establish services responsible for asking people to conduct certain activities in an urban environment in order to create data for possible policy interventions for which there does not exist useful historical data. This approach can help in addressing the challenges associated with the availability of data without raising privacy concerns, as the data generated through this approach will not represent any real individual in society.

Machine-generated artworks are now part of the contemporary art scene: they are attracting significant investments and they are presented in exhibitions together with those created by human artists. These artworks are mainly based on generative deep learning (GDL) techniques, which have seen a formidable development and remarkable refinement in the very recent years. Given the inherent characteristics of these techniques, a series of novel legal problems arise. In this article, we consider a set of key questions in the area of GDL for the arts, including the following: is it possible to use copyrighted works as training set for generative models? How do we legally store their copies in order to perform the training process? Who (if someone) will own the copyright on the generated data? We try to answer these questions considering the law in force in both the United States and the European Union, and potential future alternatives. We then extend our analysis to code generation, which is an emerging area of GDL. Finally, we also formulate a set of practical guidelines for artists and developers working on deep learning generated art, as well as some policy suggestions for policymakers.

U.S., UK, and European municipalities are increasingly experimenting with data as an instrument for social policy. This movement pertains often to the design of municipal data warehouses, dashboards, and predictive analytics, the latter mostly to identify risk of fraud. This transition to data-driven social policy, captured by the term “digital welfare state,” almost completely takes place out of political and social view, and escapes democratic decision making. In this article, I zoom in on The Netherlands and show in detail how sound data governance is lacking at three levels: data experiments and practices take place in a so-called “institutional void” without any clear democratic mandate; moreover, they are often based on disputable quality of data and analytic models; and they tend to transgress the recent EU General Data Protection Regulation (GDPR) about privacy and data protection. I also assess that key stakeholders in this data transition, that is the citizens whose data are used, are not actively informed let alone invited to participate. As a result, a practice of top-down monitoring, containment and control is evolving despite the desire of civil servants in this domain to do “good” with data. I explore several data and policy alternatives in the conclusion to contribute to a higher quality and more democratic usage of data in the digital welfare state.

Large-scale coordinated efforts have been dedicated to understanding the global health and economic implications of the COVID-19 pandemic. Yet, the rapid spread of discrimination and xenophobia against specific populations has largely been neglected. Understanding public attitudes toward migration is essential to counter discrimination against immigrants and promote social cohesion. Traditional data sources to monitor public opinion are often limited, notably due to slow collection and release activities. New forms of data, particularly from social media, can help overcome these limitations. While some bias exists, social media data are produced at an unprecedented temporal frequency, geographical granularity, are collected globally and accessible in real-time. Drawing on a data set of 30.39 million tweets and natural language processing, this article aims to measure shifts in public sentiment opinion about migration during early stages of the COVID-19 pandemic in Germany, Italy, Spain, the United Kingdom, and the United States. Results show an increase of migration-related Tweets along with COVID-19 cases during national lockdowns in all five countries. Yet, we found no evidence of a significant increase in anti-immigration sentiment, as rises in the volume of negative messages are offset by comparable increases in positive messages. Additionally, we presented evidence of growing social polarization concerning migration, showing high concentrations of strongly positive and strongly negative sentiments.

Explainability is highly desired in machine learning (ML) systems supporting high-stakes policy decisions in areas such as health, criminal justice, education, and employment. While the field of explainable ML has expanded in recent years, much of this work has not taken real-world needs into account. A majority of proposed methods are designed with generic explainability goals without well-defined use cases or intended end users and evaluated on simplified tasks, benchmark problems/datasets, or with proxy users (e.g., Amazon Mechanical Turk). We argue that these simplified evaluation settings do not capture the nuances and complexities of real-world applications. As a result, the applicability and effectiveness of this large body of theoretical and methodological work in real-world applications are unclear. In this work, we take steps toward addressing this gap for the domain of public policy. First, we identify the primary use cases of explainable ML within public policy problems. For each use case, we define the end users of explanations and the specific goals the explanations have to fulfill. Finally, we map existing work in explainable ML to these use cases, identify gaps in established capabilities, and propose research directions to fill those gaps to have a practical societal impact through ML. The contribution is (a) a methodology for explainable ML researchers to identify use cases and develop methods targeted at them and (b) using that methodology for the domain of public policy and giving an example for the researchers on developing explainable ML methods that result in real-world impact.

The COVID-19 pandemic has exposed the need for more contactless interactions, leading to an acceleration in the design, development, and deployment of digital identity tools and contact-free solutions. A potentially positive outcome of the current crisis could be the development of a more data privacy and human rights compliant framework for digital identity. However, for such a framework to thrive, two essential conditions must be met: (1) respect for and protection of data privacy irrespective of the type of architecture or technology chosen and (2) consideration of the broader impacts that digital identity can have on individuals’ human rights. The article draws on legal, technology-facing, and policy-oriented academic literature to evaluate each of these conditions. It then proposes two ways to leverage the process of digitalization strengthened by the pandemic: a data privacy-centric and a human rights-based approach to digital identity solutions fit for post-COVID-19 societies.

Health data have enormous potential to transform healthcare, health service design, research, and individual health management. However, health data collected by institutions tend to remain siloed within those institutions limiting access by other services, individuals or researchers. Further, health data generated outside health services (e.g., from wearable devices) may not be easily accessible or useable by individuals or connected to other parts of the health system. There are ongoing tensions between data protection and the use of data for the public good (e.g., research). Concurrently, there are a number of data platforms that provide ways to disrupt these traditional health data siloes, giving greater control to individuals and communities. Through four case studies, this paper explores platforms providing new ways for health data to be used for personal data sharing, self-health management, research, and clinical care. The case-studies include data platforms: PatientsLikeMe, Open Humans, Health Record Banks, and unforgettable.me. These are explored with regard to what they mean for data access, data control, and data governance. The case studies provide insight into a shift from institutional to individual data stewardship. Looking at emerging data governance models, such as data trusts and data commons, points to collective control over health data as an emerging approach to issues of data control. These shifts pose challenges as to how “traditional” health services make use of data collected on these platforms. Further, it raises broader policy questions regarding how to decide what public good data should be put towards.

Datafied societies need informed public debate about the implications of data science technologies. At present, internet users are often unaware of the potential consequences of disclosing personal data online and few citizens have the knowledge to participate in such debates. This paper argues that critical big data literacy efforts are one way to address this lack of knowledge. It draws on findings from a small qualitative investigation and discusses the effectiveness of online critical big data literacy tools. Through pre and post use testing, the short- and longer-term influence of these tools on people’s privacy attitudes and behavior was investigated. The study’s findings suggested that the tools tested had a predominantly positive initial effect, leading to improved critical big data literacy among most participants, which resulted in more privacy-sensitive attitudes and internet usage. When analyzing the tools’ longer-term influence, results were more mixed, with evidence suggesting for some that literacy effects of the tools were short-lived, while for others they led to more persistent and growing literacy. The findings confirm previous research noting the complexity of privacy attitudes and also find that resignation toward privacy is multi-faceted. Overall, this study reaffirms the importance of critical big data literacy and produces new findings about the value of interactive data literacy tools. These tools have been under-researched to date. This research shows that these tools could provide a relevant means to work toward empowering internet users, promoting a critical internet usage and, ideally, enabling more citizens to engage in public debates about changing data systems.

In this discussion paper, we outline the motivations and the main principles of the Trusted Smart Statistics (TSS) concept that is under development in the European Statistical System. TSS represents the evolution of official statistics in response to the challenges posed by the new datafied society. Taking stock from the availability of new digital data sources, new technologies, and new behaviors, statistical offices are called nowadays to rethink the way they operate in order to reassert their role in modern democratic society. The issue at stake is considerably broader and deeper than merely adapting existing processes to embrace so-called Big Data. In several aspects, such evolution entails a fundamental paradigm shift with respect to the legacy model of official statistics production based on traditional data sources, for example, in the relation between data and computation, between data collection and analysis, between methodological development and statistical production, and of course in the roles of the various stakeholders and their mutual relationships. Such complex evolution must be guided by a comprehensive system-level view based on clearly spelled design principles. In this paper, we aim at providing a general account of the TSS concept reflecting the current state of the discussion within the European Statistical System.

Data trusts have been conceived as a mechanism to enable the sharing of data across entities where other formats, such as open data or commercial agreements, are not appropriate, and make data sharing both easier and more scalable. By our definition, a data trust is a legal, technical, and organizational structure for enabling the sharing of data for a variety of purposes. The concept of the “data trust” requires further disambiguation from other facilitating structures such as data collaboratives. Irrespective of the terminology used, attempting to create trust in order to facilitate data sharing, and create benefit to individuals, groups of individuals, or society at large, requires at a minimum a process-based mechanism, that is, a workflow that should have a trustworthiness-by-design approach at its core. Data protection by design should be a key component of such an approach.

This commentary looks at the use of corporate social responsibility (CSR) mechanisms for implementing responsible data use. The commentary offers an overview of CSR theory and the discourse on a growing phenomenon known as corporate digital responsibility (CDR). The commentary links these theories to the historical debates on the nature of technology, ethics, and society. The aim is to reflect on CSR and CDR mechanisms and ignite the discussion on their adequacy considering the pursuit of data responsibility. Through our discussion and brief case studies, the paper reveals the gaps in relying on CSR and CDR and the need for a broader societal and comprehensive approach.

Like a hydra, fraudsters adapt and circumvent increasingly sophisticated barriers erected by public or private institutions. Among these institutions, banks must quickly take measures to avoid losses while guaranteeing the satisfaction of law-abiding customers. Facing an expanding flow of operations, effective banking relies on data analytics to support established risk control processes, but also on a better understanding of the underlying fraud mechanism. In addition, fraud being a criminal offence, the evidential aspect of the process must also be considered. These legal, operational, and strategic constraints lead to compromises on the means to be implemented for fraud management. This paper first focuses on the translation of practical questions raised in the banking industry at each step of the fraud management process into performance evaluation required to design a fraud detection model. Secondly, it considers a range of machine learning approaches that address these specificities: the imbalance between fraudulent and nonfraudulent operations, the lack of fully trusted labels, the concept-drift phenomenon, and the unavoidable trade-off between accuracy and interpretability of detection. This state-of-the-art review sheds some light on a technology race between black box machine learning models improved by post-hoc interpretation and intrinsic interpretable models boosted to gain accuracy. Finally, it discusses how concrete and promising hybrid approaches can provide pragmatic, short-term answers to banks and policy makers without swallowing up stakeholders with economical and ethical stakes in this technological race.

A number of governmental and nongovernmental organizations have made significant efforts to encourage the development of artificial intelligence in line with a series of aspirational concepts such as transparency, interpretability, explainability, and accountability. The difficulty at present, however, is that these concepts exist at a fairly abstract level, whereas in order for them to have the tangible effects desired they need to become more concrete and specific. This article undertakes precisely this process of concretisation, mapping how the different concepts interrelate and what in particular they each require in order to move from being high-level aspirations to detailed and enforceable requirements. We argue that the key concept in this process is accountability, since unless an entity can be held accountable for compliance with the other concepts, and indeed more generally, those concepts cannot do the work required of them. There is a variety of taxonomies of accountability in the literature. However, at the core of each account appears to be a sense of “answerability”; a need to explain or to give an account. It is this ability to call an entity to account which provides the impetus for each of the other concepts and helps us to understand what they must each require.

Today, technological developments are ever-growing yet fragmented. Alongside inconsistent digital approaches and attitudes across city administrations, such developments have made it difficult to reap the benefits of city digital twins. Bringing together experiences from five research projects, this paper discusses these digital twins based on two digital integration methodologies—systems and semantic integration. We revisit the nature of the underlying technologies, and their implications for interoperability and compatibility in the context of planning processes and smart urbanism. Semantic approaches present a new opportunity for bidirectional data flows that can inform both governance processes and technological systems to co-create, cross-pollinate, and support optimal outcomes. Building on this opportunity, we suggest that considering the technological dimension as a new addition to the trifecta of economic, environmental, and social sustainability goals that guide planning processes, can aid governments to address this conundrum of fragmentation, interoperability, and compatibility.

In our data-driven society, personal data affecting individuals as data subjects are increasingly being collected and processed by sizeable and international companies. While data protection laws and privacy technologies attempt to limit the impact of data breaches and privacy scandals, they rely on individuals having a detailed understanding of the available recourse, resulting in the responsibilization of data protection. Existing data stewardship frameworks incorporate data-protection-by-design principles but may not include data subjects in the data protection process itself, relying on supplementary legal doctrines to better enforce data protection regulations. To better protect individual autonomy over personal data, this paper proposes a data protection-focused data commons to encourage co-creation of data protection solutions and rebalance power between data subjects and data controllers. We conduct interviews with commons experts to identify the institutional barriers to creating a commons and challenges of incorporating data protection principles into a commons, encouraging participatory innovation in data governance. We find that working with stakeholders of different backgrounds can support a commons’ implementation by openly recognizing data protection limitations in laws, technologies, and policies when applied independently. We propose requirements for deploying a data protection-focused data commons by applying our findings and data protection principles such as purpose limitation and exercising data subject rights to the Institutional Analysis and Development (IAD) framework. Finally, we map the IAD framework into a commons checklist for policy-makers to accommodate co-creation and participation for all stakeholders, balancing the data protection of data subjects with opportunities for seeking value from personal data.

The rapid spread of COVID-19 infections on a global level has highlighted the need for accurate, transparent and timely information regarding collective mobility patterns to inform de-escalation strategies as well as to provide forecasting capacity for re-escalation policies aiming at addressing further waves of the virus. Such information can be extracted using aggregate anonymized data from innovative sources such as mobile positioning data. This paper presents lessons learnt and results of a unique Business-to-Government initiative between several mobile network operators in Europe and the European Commission. Mobile positioning data have supported policy-makers and practitioners with evidence and data-driven knowledge to understand and predict the spread of the disease, the effectiveness of the containment measures, their socio-economic impacts while feeding scenarios at European Union scale and in a comparable way across countries. The challenges of these data sharing initiative are not limited to data quality, harmonization, and comparability across countries, however important they are. Equally essential aspects that need to be addressed from the onset are related to data privacy, security, fundamental rights, and commercial sensitivity.

Increasingly, laws are being proposed and passed by governments around the world to regulate artificial intelligence (AI) systems implemented into the public and private sectors. Many of these regulations address the transparency of AI systems, and related citizen-aware issues like allowing individuals to have the right to an explanation about how an AI system makes a decision that impacts them. Yet, almost all AI governance documents to date have a significant drawback: they have focused on what to do (or what not to do) with respect to making AI systems transparent, but have left the brunt of the work to technologists to figure out how to build transparent systems. We fill this gap by proposing a stakeholder-first approach that assists technologists in designing transparent, regulatory-compliant systems. We also describe a real-world case study that illustrates how this approach can be used in practice.

Data governance is an emerging field of study concerned with how a range of actors can successfully manage data assets according to rules of engagement, decision rights, and accountabilities. Urban studies scholarship has continued to demonstrate and criticize lack of community engagement in smart city development and urban data governance projects, including in local sustainability initiatives. However, few move beyond critique to unpack in more detail what community engagement should look like. To overcome this gap, we develop and test a participatory methodology to identify approaches to empowering community engagement in data governance in the context of the Monash Net Zero Precinct in Melbourne, Australia. Our approach uses design for social innovation to enable a small group of “precinct citizens” to co-design prototypes and multicriteria mapping as a participatory appraisal method to open up and reveal a diversity of perspectives and uncertainties on data governance approaches. The findings reveal the importance of creating deliberative spaces for pluralising community engagement in data governance that consider the diverse values and interests of precinct citizens. This research points toward new ways to conceptualize and design enabling processes of community engagement in data governance and reflects on implementation strategies attuned to the politics of participation to support the embedding of these innovations within specific socio-institutional contexts.