Impartiality, broadly meaning the absence of bias and according equal treatment before the law, is a foundational element of judicial decision-making around the world. In this chapter, we consider how the goal of judicial impartiality may be either enhanced and supported or undermined by the use of artificial intelligence. Key developments in legal AI include innovations directed toward courts and decision- makers. These may be process-driven – for example, triaging or decision-supporting systems; in the case of pre-trial processes, judges may need to manage technology-facilitated document discovery. AI systems may also be involved in the production of evidence submitted to the court. Finally, courts and judges themselves may be the subjects of AI tools, such as those which identify patterns in decision-making. As this chapter explores, these different uses all have implications for the way that judicial impartiality is enacted and tested.

Artificial intelligence (AI) is influencing many fields around the world, including court practice and procedure. This chapter focuses on AI and the courts in Australia and New Zealand, examining both criminal and civil trial applications. The first section discusses generative AI, technology-assisted review, and automated decision- making; and the second considers the influence of AI on criminal cases, with a focus on child protection and sentencing. AI has many useful applications in this context, however, it should be carefully regulated. In relation to the development of policy and guidelines on AI, Australia and New Zealand courts are only beginning their implementation and may not be as advanced as other jurisdictions, but there is increasing recognition in government and by legal regulatory bodies, and this will be an area of significant policy development over the next decade.

This chapter explores the implications of AI for human judges through the lens of judicial professional competence. It draws on Australasian experience to make two universal arguments: to include competence on the front bench of judicial regulatory values, and to embed digital literacy in the definition and pursuit of judicial competence. There is a deep-rooted, but increasingly problematic, assumption in common law jurisdictions that judges emerge ready-made from the ranks of senior lawyers. The breadth and complexity of potential judicial engagement with AI poses a profound challenge to this assumption. Even in ‘career’ judiciaries, traditional markers of competence for judicial work do not reliably translate to competence for AI. While other dimensions of modern judicial competence, like cross-cultural skills, may be seen to raise similar concerns, AI-related risks and opportunities are proving unique in the speed at which they emerge and evolve. There is an urgent need for more open discussion about equipping future (and current) judicial cohorts to meet this challenge.

Efficiency is one of the most pervasive arguments in favour of implementing algorithms in courts of law. Across different legal contexts, many judiciaries find themselves pressured towards efficiency by growing caseloads and budgetary constraints. The purported speed of the use of AI can be seen as a solution to many existing problems, and even as a positive contribution to the value of access to justice. Through a case study of the Brazilian Judiciary’s strategy of the implementation of algorithms, the drive towards efficiency is examined and unpacked to reveal a series of tensions. First, there is a lack of conceptual clarity which leads to multiple, and sometimes competing, notions of efficiency, especially in light of the interpretation and interplay of legal principles. Moreover, the neutral appearance of efficiency can obscure political choices that cause substantive changes to the legal system without being submitted to democratic control. In this sense, a more nuanced view of efficiency as a judicial value is necessary, where it can be both contested and balanced against other core judicial values, and also seen as directional and at the service of specific ends of law.

Research on judicial use of AI has mainly focused on general attitudes toward algorithmic decision-making, leaving open the question of how policy choices shape public perceptions of the courts. This chapter addresses this gap through a comparative analysis of judicial AI policies across four major jurisdictions: the EU, UK, US, and China. We identify three key dimensions along which these approaches differ: the choice between hard and soft law, transparency requirements, and restrictions on substantive versus administrative use. Drawing on insights from rational choice theory and behavioural economics, we analyse how each regulatory choice might influence public trust and legitimacy. Our analysis suggests that the effectiveness of different approaches likely depends on institutional fit, including the pre-existing legal culture, levels of trust in courts and technology, and broader societal attitudes toward automation. These findings help explain the emergence of divergent regulatory approaches across jurisdictions and offer insights for policy-makers seeking to maintain public confidence in the courts while integrating AI into judicial systems.

There is a deep scepticism concerning the idea that AI should be used in the making of judicial decisions. There are normative risks such as inaccuracy and a lack of explainability and accountability, and there are sociological risks to public trust in the judicial system. Prominent legal instruments such as the EU AI Act, Vilnius Convention, and General Data Protection Regulation (GDPR) seek to set some clear guardrails around the use of AI in judicial decision-making, but face two problems. First, they underappreciate the Collingridge dilemma, in which premature intervention risks over-regulation, while belated intervention risks under-regulation. Second, there is a misplaced faith in the power of legal obligations to provide sufficient (and enforceable) guidance. This chapter asks what model of governance should be adopted for the use of AI in courts. In doing so, it undertakes a survey of the current status and evolution of AI technology in courts, examines how we should evaluate risks, and considers competing governance models. It argues that a model of anticipatory governance, often suitable for long and complex problems, should be adopted, and some of the implications are discussed.

Artificial intelligence (AI) is used in criminal courtrooms to support judicial decision- making. Despite heavy criticism on opacity, complexity, non-contestability, or unfair discrimination, such uses have been favoured, given AI’s promises of efficiency, effectiveness, and accuracy in the overall decision-making process. Focusing on the use of AI-generated evidence, this chapter analyses various European frameworks on evidence and fair trial scheme, the data protection guarantees under the Law Enforcement Directive (LED) and the requirements for AI use by the judiciary set out by the AI Act. We assess whether and to what degree the use of AI in criminal courtrooms can respect fundamental European principles regarding human rights and defence rights.

Adjudication in the various courts in Nigeria has been struggling to break through the jinx of case backlogs, slow legal service delivery, limited access to justice, particularly in rural areas, overburdened courts, and insufficient legal resources. However, Covid-19 led to the adoption of digital technology in the filing, service of court processes, and speedy dispensation of justice through virtual court sittings. Technologies and artificial intelligence (AI) are the driving force behind the digital administration of law and justice, and courts in Nigeria stand at the precipice of a potential technological breakthrough. However, this is not without challenges, such as inadequate infrastructure, poor funding, and lack of capacity building for the judges to understand the complexities. This chapter examines the development of the AI-driven court system and attempts to put in place a national AI policy for the justice sector in Nigeria.

The prologue fleshes out the lessons drawn from this book. It offers best practices for a workable AI governance model that uses technical solutions, business self-regulation, and legal regulation. Then, it delves into some of the shortcomings of that model. The radical-democratic perspective that I advocate makes five general, practical suggestions for everyone concerned with AI risks and harms. (1) Organize: Build networks of support and civic organizations around technology-specific concerns as well as conventional rights considerations; (2) Learn: Acquire cross-disciplinary capabilities on the uses, practical applications, potential risks, and governance models associated with technologies like AI; (3) Participate: Push politicians and businesses to expand the boundaries of decision-making in the public and private sectors; (4) Care: Approach technological change from the perspective of vulnerable populations, and with an ethic of non-domination that refuses to treat nature and other people as instruments; and (5) Resist: Maintain an openness to contention with the producers and users of technologies that generate risks and harms.

Chapter 7 zooms out of conceptual and empirical studies of AI governance to ask if we can build a better future with AI. The technical, corporate, and legal governance models presented in this book are necessary but insufficient to endow ordinary people with the power to push back against risks and harms, and chart a course for AI for the common good. Thinking together with philosophers and social scientists in the Critical Theory, Science and Technology Studies, and Democratic Theory traditions, I argue that most people’s experience with AI is one of fear as a result of their long-standing disempowerment and alienation from the technologies shaping their lives. Attributing disempowerment and alienation to technical aspects of AI is wrongheaded: It is the evolution of modern capitalism that has widened the gap between people and the technologies that are supposed to make their lives better. Reorienting the relationship between people and AI requires a radical-democratic politics that questions hierarchy in government and in the workplace. Technology can serve as a force for the social good only if informed citizens participate in the decisions shaping their lives in the design, development, deployment, and use of modern technology, AI included.

Chapter 3 presents the other side of the coin, namely AI risks and harms. Automated decision systems, chatbots, recommender systems, and other AI-powered software and platforms have been found to cause potential risks or actual harms to affected persons and communities. Such risks and harms include bias and discrimination, surveillance, inaccurate, incorrect and unreliable output, disinformation, misinformation or manipulation, harm to life, livelihood and wellbeing, privacy violations, decline in product and service quality, political polarization, online radicalization and algorithmic censorship, and job replacement. Some of these harms, such as bias and discrimination, have already been experienced frequently, while others, like job replacement, point to future risks. It is also worth noting that AI risks and harms often aggravate existing social and political problems. For example, political polarization and radicalization, while exacerbated by algorithmic curation, appear to have origins in societal divisions. Finally, AI is criticized for causing system-level harm in the form of environmental degradation, exploitation of labor, and market concentration.

Chapter 2 is devoted to AI ethics, broadly defined. It provides an overview of ethical, responsible, safe, trustworthy, transparent and explainable, accurate, just and fair, accountable, sustainable, robust, accessible and inclusive AI. Just as the definition of AI itself is fraught with disagreement, words with a connotation of “good” AI have generated considerable controversy among academics, social movement activists, journalists, business leaders, and lawmakers. This chapter aims to represent the plurality of positions. Furthermore, the adjectives associated with getting AI right are mutually supportive, but tensions between desirable goals are mentioned as well.