Digital transformation (DT) is increasingly recognized as vital to enhancing the organizational performance of public sector worldwide. Yet many DT projects fail to achieve their intended outcomes. We argue that these failures are partly attributable to the risk environments surrounding projects before and after implementation. Although DT literature emphasizes risk management and resilience, limited attention has been given to integrating both into DT project design. This paper addresses this gap by examining how these considerations are embedded as strategic elements in DT projects in the Global South. Using a convergent-parallel mixed-methods approach, we analyzed 32 World Bank Group DT project proposals from governments in the Global South using BERTopic-based natural language processing, triangulated with thematic analysis of 13 semi-structured interviews with third-party stakeholders. Findings reveal a systematic gap between resilience assumptions embedded in project designs and the contextual realities of implementation. Specifically, limited local capacity, governance fragmentation, remoteness, geopolitical conflict, and dependence on external suppliers can significantly affect project outputs and outcomes. We conclude by proposing project-level mechanisms to strengthen resilience and identify directions for future research on government-led DT projects in developing-country contexts.
Digital resilience has gained increasing attention as public organizations face accelerating digital transformation and growing social threats. However, research remains fragmented, providing limited understanding of how digital resilience develops in public service delivery and how its determinants relate to its manifestations. This study addresses this gap by conceptualizing digital resilience from an organizational perspective and examining the relationships between its determinants and manifestations in public service delivery. The analysis draws on 59 semi-structured interviews conducted in Poland, a relevant setting due to its advanced public-sector digital transformation and exposure to multiple contemporary crises. Using thematic analysis and the Gioia methodology, the study identifies three manifestations of digital resilience: continuity and availability of public services, organizational development, and technologization of internal and external processes. It further reveals four determinant groups: social, technological, organizational, and legal. The findings show how these determinants jointly enable and shape the identified manifestations, resulting in different resilience outcomes in public service delivery. Building on these relationships, the study develops a stakeholder-driven conceptual model and proposes a definition of digital resilience. The study contributes to digital governance research by offering conceptual clarity and an integrated framework that explains how digital resilience develops in public service delivery.
Local governments are beginning to use artificial intelligence (AI) in internal administrative work, including recruitment. In a decentralized municipality, however, a standardized AI assessment must travel between a technology vendor, central HR specialists, recruiters, and hiring managers responsible for different services. This creates a tension between representational portability and contextual adequacy. Based on a qualitative case study of a large Swedish municipality, we examine an AI interviewing system that translated candidates' open-ended responses into competency indicators and narrative summaries used in early-stage recruitment. We develop a process model of semantic reconciliation comprising five recurrent forms of work: scoping what could be delegated to the system, mapping vendor categories to municipal and occupational vocabularies, refining interpretations when role context was lost, embedding shared interpretations across a distributed organization, and composing AI reports with locally accountable human judgment. The study contributes to research on local-government AI and data governance by showing how semantic machines make evaluative information portable while requiring continuous organizational work to restore context. It also specifies practical arrangements through which municipalities can use common AI tools without allowing standardized outputs to displace role-specific knowledge or human decision authority.
Digital platforms are increasingly used to address regulatory fragmentation, yet evidence from actual regulatory behavior remains limited. This study examines how inspection coordination is organized under Nanjing's “181” integrated emergency management platform. Drawing on 37,876 inspection records from August 2023 to January 2025, we construct a 2-mode inspection network linking 719 inspection departments and 7934 regulated entities through 34,053 inspection ties, and estimate Exponential Random Graph Models (ERGMs). The results show that coordination is organized through three related patterns. Structurally, district-level departments occupy central positions, with the strongest cluster in Yuhuatai District. In terms of coordination coverage, inspection overlap is concentrated around selected regulated entities rather than evenly distributed, a pattern consistent with shared regulatory attention. In terms of coordination responsiveness, departments that identify hazards more often are more likely to form inspection ties. These findings show how platform-generated administrative records reveal the organization of regulatory coordination in a hierarchical administrative system through recorded activities, shared targets, and risk information.
Public agencies are strategically adapting and applying digital technologies to improve their business processes and deliver greater value to society. This development has a profound impact on the agencies' institutional arrangements. While some services that the agencies offer are almost fully automated, others still require a high level of caseworker involvement. The variance in levels of automation and the coexistence of different institutional arrangements within the same agency are interesting and challenging because they contradict predictions made by both scholars and policymakers. Therefore, we present a study of three large Scandinavian public welfare agencies, utilizing document analysis supplemented by interviews with top managers and senior IT and business developers. We apply Bovens and Zouridis' (2002) typology of street-, screen-, and system-level bureaucracies as our main analytical lens, combined with Christensen and Lægreid's (2011) concepts of hybridity and sedimentation, with a socio-technical perspective in the background. We contribute to digital government and public administration research by enhancing theoretical understanding of the relationship between digital technologies, institutional arrangements and organizational trajectories. Specifically, we identify the hybrid bureaucracy as a composite, multifunctional organizational entity that encompasses several institutional arrangements simultaneously and combines aspects of street-, screen-, and system-level bureaucracies while also exhibiting its own distinct characteristics.
Governments worldwide increasingly leverage technology to transform service delivery through digital services. While these services have been linked to gains in efficiency and quality, they are not immune to the administrative burden. This paper explores the administrative burden in digital services, addressing the research question: How does administrative burden emerge and vary across the stages of digital service delivery from the citizens' perspective? Based on data from 13 focus groups with Mexican citizens, we analyze digital services using a process model – adapted from studies of online information-seeking and electronic commerce – as our framework. Findings suggest that experiences of burden vary along the stages of the service delivery process. Our study contributes to research by examining an administrative burden in digital services through a process-specific lens. It also offers practical implications for public managers in designing inclusive digital services that respond to diverse needs and shift the burden away from the citizens.
This article investigates how Members of the European Parliament (MEPs) constructed policy narratives on artificial intelligence (AI) in the legislative process leading up to the EU Artificial Intelligence Act (AI Act) adopted in 2024. While previous research has primarily focused on executive-driven strategy documents, the Parliament's discursive role remains underexplored. To address this gap, the study applies the Narrative Policy Framework (NPF) to a corpus of 18 parliamentary documents from 2020 to 2024, in which 500 narratives were manually coded according to the NPF categories of plot, character, setting, and moral. The findings reveal remarkable stability in the Parliament's AI narratives over time, despite significant external shocks such as the release of ChatGPT 3.5 in 2022. Plots were dominated by stories of control and rising, supported by heroes and beneficiaries, framing AI as both manageable and beneficial. Settings spanned security, global competition, and European legal norms, while morals centered on risk-based regulation, ethical standards, and calls for a joint approach. By illuminating how the European Parliament narrates AI policy, this study contributes to understanding the discursive dynamics shaping EU AI governance and their implications for digital government research, particularly regarding the normative conditions under which AI systems are adopted in public administration.
Design science research (DSR) has been advocated for combining relevance and rigor and is socio-technical by nature. Yet most articles in the digital government field focus on the resulting design artifacts, presenting only methodologies and evaluations, and generating limited actionable, generalizable design knowledge. This editorial analyzes multiple approaches to advancing the development of prescriptive design principles and policy statements. The generation and rationale of prescriptive knowledge are discussed, along with various forms of design knowledge evaluation, to facilitate its connection to specific contexts. Three foundational propositions are presented to identify orienting strategies for developing prescriptive knowledge, explain its underlying rationale, and evaluate design knowledge. An integrative framework for generating and grounding prescriptive knowledge in line with the classical DSR paradigm is developed. This framework provides the basis for researchers to make generalizable knowledge contributions to the field using DSR.
Workplace surveillance is an increasingly pervasive feature of contemporary organisations, yet research overwhelmingly positions employees as its primary subjects and agents of resistance. Managers, by contrast, are implicitly framed as a homogeneous group who design, implement, and sustain surveillance. We argue that such framing ignores the complex positions managers simultaneously occupy as implementers, mediators, and potential subjects of surveillance, and further obscures situations in which they themselves are monitored, oppose surveillance, or actively engage in resisting it. Drawing on a grounded theory study of a UK local government authority, we examine how and why managers resist digitally augmented surveillance systems, revealing three interrelated resistance strategies: shielding employees from surveillance, obfuscating data to undermine system integrity, and deploying discursive practices to contest surveillance legitimacy. These strategies are simultaneously shaped by self-protective and value-driven motivations. By integrating sociotechnical systems theory with public service ethics, we develop a conceptual framework that repositions managers as agentive, morally reasoning actors navigating the contested terrain of digital control. The findings advance interdisciplinary understanding of resistance and ethics in digitally mediated public sector organisations, and respond to calls for research that recognises managerial diversity and conceptualises resistance as a situated sociotechnical process.
Public corruption undermines government performance, erodes trust, and fuels illiberal populism. While macro-level drivers of corruption are well documented, little is known about the individual-level attitudes and beliefs associated with public servants' susceptibility to engaging in corrupt behaviour. Using a large-scale, globally stratified dataset of 18,277 public servants across 90 countries, we systematically compare a conventional regression baseline with five contemporary supervised machine learning models to predict susceptibility to corruption. Machine learning improves predictive accuracy and shows that democratic values, beliefs about competition, and attitudes towards leadership are more consistently predictive than socio-economic characteristics such as income, education, or gender. These patterns are robust across modelling approaches and analytical settings. The findings illustrate the value of transparent model comparison and machine learning for studying integrity risks, provide exploratory insights for corruption prevention strategies, and open new avenues for research on predictive modelling for public sector integrity.
Policy documents guide technological transformations. Prior research on national artificial intelligence (AI) strategies has examined how these texts articulate governance values and principles, but how they promote identification, urgency, and action is less well understood. This gap is especially striking given arguments that national AI strategies are typically written for policy professionals. We show that these documents are also widely used as publicly accessible reference points in debates on responsible AI. Building on the narrative turn in public policy scholarship, we analyse 22 European national AI strategies. We find that AI strategies create normative entrainment through affect, constituted by three narratives: (1) establishing normative orientations by ascribing societal values to AI; (2) anchoring these orientations by invoking social emotions that embed AI in cultural identity and everyday life; and (3) promoting implementation by juxtaposing moral emotions such as fear and hope while mobilising urgency, leadership, and responsibility. By specifying how emotional narratives are woven into policy texts, this framework contributes to research on the communicative construction of AI policy and offers implications for responsible digital transformation.
Digital maturity frameworks have become central instruments of digital government practice, used by public sector organizations, national agencies, and international bodies to assess and steer digital transformation. Despite their widespread use, few frameworks have been subjected to rigorous psychometric testing. The scores they produce therefore rest on unverified assumptions about dimensionality, reliability, and validity. We address this gap by applying scale development methodology to a digital maturity framework deployed across more than 160 Swedish public sector organizations since 2018. Through exploratory and confirmatory factor analysis on a cleaned and multiply imputed sample from 118 organizations, we evaluate the framework's latent factor structure. The analysis yields a four-factor model that is empirically supported within the studied sample, comprising Digital Operations, Digital Legacy, Benefit Management, and Digital Development, subsumed under a second-order factor of Evaluated Digital Maturity. The results also reveal specific weaknesses that are predictable consequences of deploying an instrument developed without scale development methodology. We make two contributions. First, we provide a large-scale empirical test of a widely used digital maturity framework in the public sector, identifying its strengths and the implications of developing an instrument without psychometric grounding. Second, we adapt scale development methodology from prospective scale construction to retrospective framework evaluation, strengthening measurement in digital government research.
Technology decentralisation is increasingly proposed as a key feature to build more trustworthy, accessible, and innovative digital public infrastructure, yet there is limited empirical knowledge of the actual benefits that such decentralised approaches would create from a public sector perspective. Furthermore, existing literature often relies on a linear dichotomy between data supply and demand that fails to capture the complexity of decentralised data ecosystems. This paper addresses these gaps by adopting an assemblage thinking perspective to conceptualise data platforms as complex socio-technical arrangements, and by developing a public values framework to broaden the understanding of the various outcome that data platforms can create. We apply this approach to an exploratory case study of Hamburg's Urban Data Platform (UDP). Our findings demonstrate that public value creation is not determined by technical decentralisation alone but by specific architecturegovernance configurations. We illustrate that both decentral and central practices can co-occur within the same system, where the role of a leading orchestrator is crucial to drive public value creation.
Applying Artificial Intelligence (AI) to fraud detection in Public Procurement (PP) has become increasingly relevant due to its ability to process large volumes of data and identify suspicious patterns more efficiently than traditional methods. However, its implementation raises critical ethical concerns regarding respect for human autonomy, prevention of harm, fairness, and explainability. This study investigates these concerns through a four-phase methodology: (i) selection of ethical principles for AI in PP fraud detection, (ii) a Systematic Literature Review (SLR) on current approaches, (iii) cross-analysis and expert validation with senior auditors, and (iv) the construction of PRO-Trust, an ethics-oriented pipeline for constructing trustworthy AI-based fraud detection systems for PPs, based on the findings of the three former stages. PRO-Trust addresses five recurrent ethical challenges - system opacity, lack of user transparency, limitations in explainability tools, lack of operational oversight, and lack of strategic control - connected between tasks we specified for the construction process of AI-based technologies to fraud detection in PP, the responsible actors, and technological strategies to operationalize ethical principles. By doing so, this study contributes to the development of more trustworthy, transparent, and accountable AI-based fraud detection systems in public administration.
Citizen perceptions towards government social media use have been usually studied taking into consideration the impacts of socio-demographics, attitudinal or technological predispositions. However, recent developments on behavioral public administration call for integrating micro-level characteristics into research, considering personality might influence how public organizations communications and messaging are perceived. In this article, we study what factors are behind citizen-given importance towards public administration social media use, by focusing on personality traits. To approach personality traits we rely on the big five personality model and its core dimensions: extraversion, agreeableness, conscientiousness, neuroticism, and openness. Methodologically, we resorted to the Spanish case, conducting an original survey targeting citizens living in the largest municipalities. Results show introversion, agreeableness, conscientiousness and openness making citizens give more importance to public administration social media use. However, neuroticism and extraversion seem to make citizens place less importance, particularly towards the promotion of participation and collaboration through social media. Age, gender and trust in government moderated the effects of some of these traits. These results might help public administrations better understand their audiences by integrating into their monitoring practices the prediction of personality traits from citizens' digital footprints. This can be further used to personalize communications and services.
Many technology projects in government are developed and implemented with third party organizations through public-private partnerships. This project studies how e-government collaborations between third parties and government affect people's trust in government and uptake of online public services in Kenya. We use an endorsement-style survey experiment to study whether trust can be transferred from a trusted company to the government. Randomizing respondents into learning about a new online public service portal delivered by 1) a firm, 2) government, or 3) a partnership between the firm and government, we find that learning about the partnership does not increase respondents' trust in government and thus do not find support for trust transfer. However, we do find support for other pathways where this online portal impacted trust in government. Specifically, we find a substitution effect, where respondents are more willing to use the online portal when they learn it is delivered by the company, but also trust the government less. We also find that respondents' preexisting levels of trust in government lead to differing results, where the substitution effect seems to be driven by those who already have low trust in government at baseline. This study helps us understand conditions under which trust transfer is more likely to occur and highlights other ways trust can be affected through digitization efforts.
This study investigates citizen preferences for the design of artificial intelligence (AI) in public administration and citizen legitimacy perceptions of AI-based tax-decisions. Analysing citizens’ perceptions of AI is important because the literature remains inconclusive and negative perceptions may undermine AI’s performance benefits. Prior research suggests that these perceptions depend on contextual, individual-level, and design-related factors. Empirically, we investigate the use of AI in taxation in Finland, a country context with high institutional trust and established use of automated decision making in taxation. Using two separate experiments with a conjoint design and a vignette design drawing on a randomly recruited citizen sample (N=1072), we offer robust evidence on the relative importance of AI decision-making characteristics and investigate perceptions of individual-level AI tax decisions. The results indicate that citizens are sceptical of AI in taxation when AI systems are granted decision-making discretion. Perceptions are further undermined when private companies are involved in developing public sector AI or when private data are used. The results have implications for the design of public sector AI, highlighting which features should be considered to ensure its use is perceived as legitimate by citizens.
Governments around the world are actively exploring how they can tackle societal problems through data-driven decision-making. As much data is now in the hands of the private sector, governments increasingly resort to purchasing data from private sources. There is, however, scant empirical evidence and a lack of understanding of the experiences of governments with this practice. This study therefore asks the following questions: how do governments purchase data from the private sector and what challenges do they encounter in the process? We conducted an exploratory multiple case study based on sixteen interviews with seven governments in the Netherlands focusing on use cases in the mobility or spatial domains. Our analysis covers, among other things, the purpose of the purchases and to what extent alternatives and joint procurement were considered, the purchasing procedures and respective market characteristics, and the governments' assessment and satisfaction with the purchases. We identify a number of challenges that are associated with data procurement. Our conclusions highlight three priorities to strengthen the position of governments as buyers of data: the need to balance competing priorities, the need to understand how dependency and lock-in emerges, and the need to expand assessments to public values.
Public organisations are struggling with the urgent need to increase their cybersecurity, i.e. the protection of information and information systems. The aim of this article is to examine how cybersecurity maturity in local government is improved through internal capacity-building and inter-municipal collaboration (IMC), and how the effectiveness of these strategies depends on the timing of their deployment in the development of organisational capacity. Using a two-wave survey of Swedish municipalities (2019 and 2023), we find that internal capacity-building (operationalised as the time allocated to Chief Information Security Officers, CISOs) and IMC through the Swedish Association of Local Authorities and Regions (SALAR) are associated with improvements in cybersecurity maturity, but collaboration with neighbouring municipalities is not. Moreover, the timing of these factors matters. Internal capacity-building through increasing the time allocated to the CISO is key at an early stage, while IMC through SALAR becomes more important as cybersecurity improves. Our work suggests that the level of maturity of cybersecurity capacity moderates the effect of collaboration on further improvement of cybersecurity, as higher cybersecurity capacity raises the ability to identify the problems at hand, which information from external actors is relevant and how this information should best be applied to meet local needs.
Governments increasingly engineer institutional roles dedicated to support the effective digital transformation of government and its interactions with citizens. How these roles and related digital policy advocacy evolve are important aspects of digital government. Our study specifically focuses on the role of narrative work in the establishment of the Digital Minister role and related policy advocacy. We examine a case where a governmental “outsider,” previously a tech activist, was brought into national government in the official capacity of Digital Minister to advance an agenda of digital transformation and transparency. Using six years of government archival data, news articles, and publicly available archival interview transcripts, we theorize how institutional actors can enact narrative work that strengthens their legitimacy through the intertwined evolution of the role and the enactment of digital policy advocacy. Implications for research and practice are discussed.
