AI Regulation Research Articles

"You Can either Blame Technology or Blame a Person..." --- A Conceptual Model of Users' AI-Risk Perception as a Tool for HCI

AI-powered systems pose unknown challenges for designers, policymakers, and users, making it more difficult to assess potential harms and outcomes. Although understanding risks is a requirement for building trust in technology, users are often excluded from risk assessments and explanations in policy and design. To address this issue, we conducted three workshops with 18 participants and discussed the EU AI Act, which is the European proposal for a legal framework for AI regulation. Based on results of these workshops, we propose a user-centered conceptual model with five risk dimensions (Design and Development, Operational, Distributive, Individual, and Societal) that includes 17 key risks. We further identify six criteria for categorizing use cases. Our conceptual model (1) contributes to responsible design discourses by connecting the risk assessment theories with user-centered approaches, and (2) supports designers and policymakers in more strongly considering a user perspective that complements their own expert views.

Truly Risk-based Regulation of Artificial Intelligence How to Implement the EU’s AI Act

Abstract The Artificial Intelligence Act (AI Act) of the European Union (EU) claims to be based on a risk-based approach to avoid over-regulation and to respect the principle of legislative proportionality. This paper argues that risk-based regulation is indeed the right approach to AI regulation. At the same time, however, the paper shows that important provisions of the AI Act do not follow a truly risk-based approach. Yet, this is nothing that cannot be fixed. The AI Act provides for sufficient tools to support future-proof legislation and to implement it in line with a genuine risk-based approach. Against this background, the paper analyses how the AI Act should be applied and implemented according to its original intention of a risk-based approach, and what lessons legislators around the world can learn from the AI Act in regulating AI.

Postphenomenological Guide to AI Regulation

Postphenomenological Guide to AI Regulation

Ethics at the heart of AI regulation

Ethics at the heart of AI regulation

A Forum on Workplace AI Regulation Around the World

A Forum on Workplace AI Regulation Around the World

Regulating the ‘AI Leviathan’: A Framework for AI Surveillance Disclosure Reporting

In an era marked by the ubiquity of digital surveillance, the role of AI algorithm companies has gained unparalleled prominence, especially as governments around the globe grapple with crises like the COVID-19 pandemic. In this evolving landscape, AI algorithm companies and governmental agencies collaboratively forge an AI-surveillance ecosystem, which could be termed an ‘AI Leviathan’. Within this system, the government operates in a multifaceted role while the AI companies act as key technological enablers. Despite recent regulatory efforts like the EU AI Act, which introduces a risk-based approach to AI regulation, extant legal frameworks still fall short of addressing the full spectrum of challenges that arise, including but not limited to privacy issues and algorithmic discrimination. Furthermore, the academic discourse has primarily centred on the state’s involvement in surveillance, frequently overlooking the pivotal role that AI algorithm companies play in influencing this extensive system. This article seeks to address this gap by providing a comprehensive analysis of the contribution of AI algorithm companies in facilitating governmentdriven surveillance, particularly emphasizing the Chinese scenario. Additionally, it proposes a comply-or-explain framework for AI Surveillance Disclosure Reporting, specifically designed for these companies, aiming to amplify their transparency and accountability.

Workplace AI Regulation and Worker Resistance in Korea

Workplace AI Regulation and Worker Resistance in Korea

On Using Self-Report Studies to Analyze Language Models

We are at a curious point in time where our ability to build language models (LMs) has outpaced our ability to analyze them. We do not really know how to reliably determine their capabilities, biases, dangers, knowledge, and so on. The benchmarks we have are often overly specific, do not generalize well, and are susceptible to data leakage. Recently, I have noticed a trend of using self-report studies, such as various polls and questionnaires originally designed for humans, to analyze the properties of LMs. I think that this approach can easily lead to false results, which can be quite dangerous considering the current discussions on AI safety, governance, and regulation. To illustrate my point, I will delve deeper into several papers that employ self-report methodologies and I will try to highlight some of their weaknesses.

Supporting Trustworthy AI Through Machine Unlearning

Machine unlearning (MU) is often analyzed in terms of how it can facilitate the “right to be forgotten.” In this commentary, we show that MU can support the OECD’s five principles for trustworthy AI, which are influencing AI development and regulation worldwide. This makes it a promising tool to translate AI principles into practice. We also argue that the implementation of MU is not without ethical risks. To address these concerns and amplify the positive impact of MU, we offer policy recommendations across six categories to encourage the research and uptake of this potentially highly influential new technology.

Artificial Intelligence and International Law: The Impact of Emerging Technologies on the Global Legal System

With the rapid development of artificial intelligence technology, its impact on international law is becoming increasingly significant. This paper explores the application of AI in international relations and the challenges it poses to existing legal systems, analyzing AI regulation within the current international law framework and the exploration of emerging international legal norms. Through specific cases, the paper elaborates on legal issues in the fields of intelligent weapons, cybersecurity, and data protection, emphasizing the importance of international cooperation and global governance mechanisms in addressing these challenges. Finally, the paper looks forward to future trends in AI technology and international law’s response strategies, proposing the necessity of constructing effective legal frameworks and ethical guidelines to promote the sustainable development of AI technology and the healthy development of the global legal system.

Digital Sustainable Growth Model (DSGM): Achieving Synergy Between Economy and Technology to Mitigate AGI Risks and Address Global Debt Challenges

The emergence of Artificial General Intelligence (AGI) as Artificial Humanity (AH) marks a crucial turning point in human history, influencing both the digital economy and sustainable development. AGI has the potential to either drive unprecedented progress or threaten sustainability. This perspective study critiques existing capitalist systems and proposes a shift towards sustainable digital strategies to address the uncertain future role of AGI. It aims to encourage further research and collaboration for a smooth transition to a society that integrates AGI, highlighting AGI's potential to contribute to sustainable development. The study employs a multi-step methodology, including interviews with AI models such as ChatGPT-4 and Gemini, validation of insights, and comparisons with expert reports. It introduces the Digital Sustainable Growth Model (DSGM) as a framework for harmonizing humanity with AH, providing new opportunities for growth and ethical governance. The DSGM addresses both human vulnerabilities and AH’s potential. Additionally, the study highlights the BankRabbna application as an innovative digital tool, intended to be the first of its kind—a global company primarily owned by people worldwide. Its main function is to automate AI regulation, ensuring AGI safety while also tackling global debt.

General-purpose AI regulation and the European Union AI Act

General-purpose AI regulation and the European Union AI Act

Technologies Versus Justice: Challenges of AI Regulation in the Judicial System

The article examines issues of using artificial intelligence in such a sensitive area of human activity as justice. The authors refer to numerous facts on attempts to create a kind of “smart court” in various countries. At the same time, these attempts run up against circumstances that indicate the need to establish legal restrictions on the use of artificial intelligence in the administration of justice. Moreover, according to the authors’ reasoned conviction, there are areas in which the robot judge turns out to be powerless to replace human intelligence. Based on the philosophical and legal approach to assessing such a phenomenon as digitalization and the phenomenology of legal judgment, the authors conclude that the adoption of a court decision that meets the requirements of the principle of justice is something beyond the reach of artificial intelligence. Such a decision can only be made by a human judge, but not by a robot. AI systems in the judicial system should support rather than supersede judges.

AI governance in Asia: policies, praxis and approaches

ABSTRACT This article surveys the status quo of AI readiness and governance in Asia and identify Asian approaches of doing AI regulation and governance through policy and document analysis. We note that some Asian countries are moving from ‘soft regulation’ through strategies and guidelines to ‘hard regulation’ through rule-setting and laws on AI. We argue that their AI governance approaches are greatly influenced by the existing internet governance frameworks and suggest the importance of historical understandings of Internet, telecommunications, and digital technology and governance to identify the connections and influences – and ‘path dependency’ of past policies upon their current AI strategies and governance. We anticipate that the AI regulatory landscape in Asia will become a diverse and contentious space due to global AI competition among the EU, China and the U.S. as well as the pragmatic paths that many Asian countries may take considering their own histories, economies and politics.

El Reglamento de Inteligencia Artificial de la Unión Europea: regulación de riesgos y sistemas de estandarización

AI has been consolidated in recent years as a central line of action in the strategy of public administrations, and the AI Regulation approved by the EU intends to become a pioneering global standard for the regulation of this technology. This article analyses the nature and main characteristics of this regulation approved on 13 March 2024 by the European Parliament. The study conceptualises the EU Regulation as part of a normative trend, existing both at European and international level, consisting of the regulation of risks and situations of uncertainty. On this premise, the paper analyses the legal architecture of the norm, which revolves around a conformity assessment system that relies, in its turn, on three elements: (1) risk management instruments applicable to AI developers and implementers, (2) a procedure for the development of technical standards that will determine the specific content of the obligations of the regulation, and (3) the creation of a governance system for the implementation and verification of such obligations.

UK and EU approaches to AI regulation: a world apart?

UK and EU approaches to AI regulation: a world apart?

ChatGPT for good? Taking ‘beneficence’ seriously in the regulation of generative artificial intelligence

ABSTRACT Generative AI platforms such as ChatGPT have found prominence in recent times with their ability to generate texts, images, etc. Several questions pertaining to ethical and legal issues surrounding ChatGPT have arisen. In this paper, I discuss the nature and background of generative AI, situating its development in the historical context of AI. I then discuss my primary research questions: is sufficient attention paid in the literature on ethics of AI to the principle of beneficence and is there theoretical clarity on its meaning? I highlight that while there is great deal of discussion on what harms can arise from generative AI and how to stop them, there is very little discussion on what amounts to AI-for-good, particularly in the literature of AI ethics and regulation. To the extent that such discussion exists, it pushes ahead with suggesting specific solutions, without fully addressing the underlying question of what makes their prescriptions beneficent. I demonstrate that the principle of beneficence as understood in biomedical ethics and human rights frameworks has limited utility for the ethics of generative AI. These gaps in the AI regulation are prominent, as they can derail long term progress of generative AI and the realization of its full potential.

Do Cases Generate Bad AI Law?

There’s an AI governance problem, but it’s not (just) the one you think. The problem is that our judicial system is already regulating the deployment of AI systems—yet we are not coding what is happening in the courts as privately driven AI regulation. That’s a mistake. AI lawsuits here and now are determining who gets to seek redress for AI injuries; when and where emerging claims are resolved; what is understood as a cognizable AI harm (and what is not), and why that is so. This Essay exposes how our judicial system is regulating AI today and critically assesses the governance stakes. When we do not adequately recognize how the generative AI cases being decided by today’s judges are already operating as a type of AI regulation, we fail to consider which emerging tendencies of adjudication about AI are likely to make good or bad AI law. For instance, litigation may do good agenda-setting and deliberative work as well as surface important information about the operation of private AI systems. But adjudication of AI issues can be bad too, given the risk of overgeneralization from particularized facts; the potential for too much homogeneity in the location of lawsuits and the kinds of litigants; and the existence of fundamental tensions between social concerns and current legal precedents. If we overlook these dynamics, we risk missing a vital lesson: AI governance requires better accounting for the interactive relationship between regulation of AI through the judicial system and more traditional public regulation of AI. Shifting our perspective creates space to consider new AI governance possibilities. For instance, litigation incentives (such as motivations for bringing a lawsuit or motivations to settle) or the types of remedies available may open up or close down further regulatory development. This shift in perspective also allows us to see how considerations that on their face have nothing to do with AI—such as access to justice measures and the role of judicial minimalism—in fact shape the path of AI regulation through the courts. Today’s AI lawsuits provide an early opportunity to expand AI governance toolkits and to understand AI adjudication and public regulation as complementary regulatory approaches. We should not throw away our shot.

Eu Ai Act and Its Relationship with Vietnamese Lawin Creating a Legal Policy for Ai Regulation

Eu Ai Act and Its Relationship with Vietnamese Lawin Creating a Legal Policy for Ai Regulation

AI, Law and beyond. A transdisciplinary ecosystem for the future of AI & Law

AbstractWe live in exciting times for AI and Law: technical developments are moving at a breakneck pace, and at the same time, the call for more robust AI governance and regulation grows stronger. How should we as an AI & Law community navigate these dramatic developments and claims? In this Presidential Address, I present my ideas for a way forward: researching, developing and evaluating real AI systems for the legal field with researchers from AI, Law and beyond. I will demonstrate how we at the Netherlands National Police Lab AI are developing responsible AI by combining insights from different disciplines, and how this connects to the future of our field.