The intense attention to how AI technologies (AIT) are increasingly permeating aspects of society as well as people’s everyday lives often ends in a call for ethics. The need for ethical AI, or responsible AI, or trustworthy AI is put forward by scholars as well as policymakers as the answer to the potential risks that the implementation of different forms of AITs could bring. This chapter has the ambition of bringing together the state-of-the art regarding the discussion on AI and ethics in policy – including a focus on the growing strand of critical studies on governing and policy. In doing this, the chapter also demonstrates the analytical merit of a governmentality framework when focusing on the political implications of the ethics discourse. Hence, the chapter focuses on what ethics ‘does’ with and in policy, and what the political implications of these ‘doings’ are. 

This article discusses the perspectives of European Union (EU) / European Economic Area Data Protection Authorities (DPAs) on their role in protecting democratic rights and freedoms in digitalised societies. Data Protection Authorities, which are independent regulators, are responsible for implementing the EU's General Data Protection Regulation in their respective countries. The views of DPAs are important given their special role in monitoring newly emerging digital technologies and how their use may impact on the functioning of democracies. The article highlights three key themes which emerged in interviews with 18 DPAs in answer to the question about what they consider to be the greatest challenges to democratic freedoms. These are: (1) threats to elections due to the manipulation of voters; (2) discriminatory effects of automated decision-making; and (3) broader chilling effects on democratic norms due to ubiquitous surveillance. The article then analyses the solutions named by DPAs to mitigate these challenges to identify their governing, or political, rationalities. The paper finds that the solutions available to DPAs to manage democratic harms tend to emphasise individual over collective responsibility and are connected to broader currents of neoliberal governance. The paper highlights the ways in which some DPAs act as important critical voices within their respective jurisdictions to draw political attention to potentially anti-democratic effects of certain practices, such as profiling, or to the model of digitalisation as it is currently constructed.

In an era of rapid digital transformation, we are increasingly subject to surveillance practices, despite acknowledgement of their anti-democratic effects. How is it that surveillance practices, once condemned as corrosive to democratic rights and freedoms, are now prevalent in democratic states? 

This thesis explores this question by asking how surveillance practices are governed in regulation, policy and practice. Drawing on governmentality studies and using Carol Bacchi’s WPR (What’s the Problem Represented to Be?) approach to policy analysis, the thesis identifies ‘problem representations’ in EU/EEA digitalisation and data protection regulation. The thesis finds that data protection and digitalisation policies are shaped by neoliberal rationalities, often prioritising economic growth over fundamental rights. Moreover, surveillance practices are normalised through risk-based regulation, which legitimises practices such as profiling and scoring when deemed to have met a threshold of ‘trustworthiness’. 

To highlight the contingency of our present model of digitalisation, the thesis traces representations of surveillance practices over time, highlighting a shift from outright rejection of surveillance practices as anti-democratic to a nuanced acceptance, re-branded in terms of economic progress, public interest and individual customisation. The thesis challenges the perceived neutrality and inevitability of surveillance technologies coupled to our current model of digitalisation, opening up possibilities for resistance and alternative models.

I en tid av snabb digital omställning utsätts vi i allt högre grad för olika former av övervakning, trots att vi är medvetna om dess antidemokratiska effekter. Hur kommer det sig att övervakning, som brukade anses som något skadligt för demokratiska fri- och rättigheter, nu har blivit någonting accepterat i demokratiska länder?

Denna avhandling utforskar den här frågan genom att undersöka hur övervakningsmetoder hanteras i demokratiska sammanhang. Med utgångspunkt i governmentalityforskning och Carol Bacchis WPR-approach för policyanalys, identifierar avhandlingen problemrepresentationer i EU/EES-digitalisering och dataskyddsreglering. En av avhandlingens huvudsakliga slutsatser är att digitaliserings- och dataskyddspolitiken är formad av nyliberala rationaliteter och prioriterar ekonomisk tillväxt framför grundläggande rättigheter. Dessutom normaliseras en övervakningspraxis genom riskbaserad reglering, och praktiker som profilering, meritering och poängsättning baserad på socialt beteende normaliseras och hanteras som tillförlitliga.

Analysen identifierar representationer av övervakning genom tiderna och spårar en förskjutning från att avvisa övervakningsmetoder som varandes antidemokratiska, till en nyansering och acceptans av övervakning utifrån argument om ekonomiska fördelar, allmänintresse och individuellt anpassade tjänster. Avhandlingen utmanar den upplevda neutraliteten och oundvikligheten hos dagens övervakningsteknologier och nuvarande modell för digitalisering, vilket öppnar upp för möjligheter till motstånd och alternativa lösningar och tillvägagångssätt.

In democratic states, mass surveillance is typically associated with totalitarianism. Surveillance practices more limited in their scope draw criticism for their potential to undermine democratic rights and freedoms and the functioning of representative democracies. Despite this, citizens living in political systems classed as democratic are increasingly subject to surveillance practices by both businesses and governments. This paper presents the results of a genealogy of OECD digitalisation discourse from the 1970s to the present to show how both harms and benefits of surveillance practices have been problematised. It shows how practices once considered unacceptable are increasingly portrayed as neutral, or even positive. A shift is identified from general agreement over the incompatibility of surveillance practices with democracy to greater acceptance of those practices when rebranded as tools to promote customisation, economic growth or public health. This transformation is significant because it: (1) shows the inherent instability of policies anchored to seemingly fixed or self-evident concepts such as ‘well-being’ or ‘public interest’; (2) highlights the fragility of democratic systems when things deemed harmful to their operation can be repurposed and subsequently permitted; and (3) highlights the contingency of (seemingly inevitable) surveillance practices, thereby opening up a space in which to challenge them.

How we choose to utilize digital technology has the potential to undermine the healthy functioning of democratic systems. Surveillance practices such as the tracking, collection and profiling of our online and real-world behavior pose a direct challenge to privacy rights and democratic freedoms such as fairness and anti-discrimination. This paper aims to understand how the GDPR represents risk and, in turn, how that representation shapes protection. Using Carol Bacchi's 'What's the Problem Represented to Be?' (WPR) approach to policy analysis, we illustrate how the GDPR's dual aims of protecting both people and the free flow of personal data exist in a state of tension and that the GDPR's framing of 'public interest' privileges economic growth over individual rights. Also problematic is the assumption that people are sufficiently informed to exercise control over their data, yet are being asked to agree to practices which may undermine that very autonomy.