Varieties of AI Explanations Under the Law. From the GDPR to the AIA, and Beyond

Hacker, Philipp; Passoth, Jan-Hendrik

doi:10.1007/978-3-031-04083-2_17

Cited by 19 publications

(24 citation statements)

References 64 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The AI & law researchers seem to prefer the latter solution (Pasquale 2017;Selbst and Barocas 2018;Mittelstadt et al 2019;) and offer its variants (Ye et al 2018;Prakken 2020;Prakken and Ratsma 2021). This line of research also includes numerous papers interpreting the existing (or, as in the case of EU AI Act, pending) legal requirements on explainability, criticizing them or proposing new mechanisms and provisions (Goodman and Flaxman 2017;Malgieri and Comandé 2017;Wachter et al 2017a;Selbst and Powles 2018;Casey et al 2019;Zuiderveen Borgesius 2020;Grochowski et al 2021;Kaminski 2021;Hacker and Passoth 2022;Sovrano et al 2022). It is worth noting that de lege lata objections state that the rules are vague, too weak or incompatible with AI conceptual grid rather than unnecessary.…”

Section: Discussionmentioning

confidence: 99%

The black box problem revisited. Real and imaginary challenges for automated legal decision making

et al. 2023

View full text Add to dashboard Cite

This paper addresses the black-box problem in artificial intelligence (AI), and the related problem of explainability of AI in the legal context. We argue, first, that the black box problem is, in fact, a superficial one as it results from an overlap of four different – albeit interconnected – issues: the opacity problem, the strangeness problem, the unpredictability problem, and the justification problem. Thus, we propose a framework for discussing both the black box problem and the explainability of AI. We argue further that contrary to often defended claims the opacity issue is not a genuine problem. We also dismiss the justification problem. Further, we describe the tensions involved in the strangeness and unpredictability problems and suggest some ways to alleviate them.

show abstract

Section: Discussionmentioning

confidence: 99%

The black box problem revisited. Real and imaginary challenges for automated legal decision making

et al. 2023

View full text Add to dashboard Cite

show abstract

“…Furthermore, citizens may also actively contribute to the design of novel solutions, for example in the realm of explanations regarding AI systems or justification structures. Researchers have already pointed to the advantages of using co-design strategies to this end (Liegl et al, 2015), also in AI regulation (Aldewereld & Mioch, 2021;Hacker & Passoth, 2022). Such participatory strategies may then be fused with a legal system willing and able to receive such input and accommodate temporality.…”

Section: Participation Legal Issues and Temporalitymentioning

confidence: 99%

“…9), but ignores some of the most troubling current applications. An example of this is emotion recognition systems (see Mc-Stay, 2016Hacker, 2022). Paradoxically, there seems to be too much future and too little present in Art.…”

Section: Temporality In the Aiamentioning

confidence: 99%

“…11 AIA); the transparency and the performance quality provisions (Art. 13 and 15 AIA) (on these, in particular, see Veale & Zuiderveen Borgesius, 2021;Ebers et al, 2021;Hacker & Passoth, 2022). Hence, the law compels developers to take possible future uses and concomitant risks into account when designing systems which may be used in such contexts.…”

Section: (3) Aiamentioning

confidence: 99%

See 1 more Smart Citation

Substantively smart cities – Participation, fundamental rights and temporality

Hacker¹,

Neyer²

2023

Internet Policy Review

View full text Add to dashboard Cite

Smart city applications are increasingly deployed in urban spaces around the world. We contend that they only merit the attribute "smart" if they embody what we term "substantial smartness". To develop this concept, we draw on both political and legal theories to show that citizen participation and activation, as well as respect for human and fundamental rights, are two essential dimensions of substantial smartness. Both dimensions, however, need to accommodate temporality, i.e., rapid changes in deployed technologies, their purposes and citizens' use of public infrastructure. By highlighting three examples and discussing smart city challenges to the GDPR, non-discrimination law and the proposed EU AI Act, we demonstrate that politics needs the lawand vice versa -to unlock the potential of substantively smart cities. Issue 1This paper is part of Future-proofing the city: A human rights-based approach to governing algorithmic, biometric and smart city technologies, a special issue of Internet Policy Review guest-edited by Alina Wernick and Anna Artyushina. Introduction: Interdisciplinary dynamics of smart environmentsLaw and politics relate to, and seek to govern, a world of change. Therefore, they are confronted with the challenge of adapting their content and effects over time, and mitigating risks not contemplated at the moment when decisions were made. This disconnect infuses an unavoidable element of temporality into the politico-legal realm. Such dynamics are particularly present in digital information technology, where advances in computer science and novel connections between different technical tools create a protean landscape of ever-changing and increasingly integrated data sets, models and applications. Yet, similarly, political action occurs in this space and may harness digital technology to adapt to, and contest, new practices and risks to human rights. Such dynamic, politico-legal spaces are epitomised by the so-called "smart city", marked by a constant change in technology and its purpose, and simultaneously by frequent alterations to the use of public infrastructure. The street conceived as a way of transport may suddenly become an avenue of protest. Public space in the city, instrumentally used for commerce, recreation, or movement, may thus be rapidly converted into a space of contestation (Colomb & Novy, 2016). Political protest, after all, typically converges toward cities, where political elites are located, who are often the ultimate audience of protest movements. Conversely, data collected in public spaces is prone to function creep, i.e., fast shifts in the exploitation and purpose of processing (cf. Graham & Wood, 2003;Koops, 2021). The growing deployment of data-collecting sensors in cities, coupled with analytics frameworks, thus has potentially far-reaching consequences. Public infrastructure is unavoidable for city dwellers who need to participate in many essential activities, from shopping to transport (Eckhoff & Wagner, 2017; Kroll, 2022, p. 2). That necessity highlights the need t...

show abstract

“…In this vein, growing attention has been redirected toward AI policy regulation to balance the rule of law and AI-driven business innovation [11]. Such attention denoted how ensuring compliance over explanation generation is not just a technical, yet a legal and political endeavor [12].…”

Section: Introductionmentioning

confidence: 99%

Operationalizing Explainable Artificial Intelligence in the European Union Regulatory Ecosystem

Nannini,

Alonso-Moral,

Catalá

et al. 2024

IEEE Intell. Syst.

View full text Add to dashboard Cite

The European Union's regulatory ecosystem presents challenges balancing legal and sociotechnical drivers for explainable AI systems. Core tensions emerge on dimensions of oversight, user needs and litigation. This paper maps provisions on algorithmic transparency and explainability across major EU data, AI, and platform policies using qualitative analysis. We characterize involved stakeholders and organizational implementation targets. Constraints become visible between useful transparency for accountability versus confidentiality protections. Through an AI hiring system example, we explore complications operationalizing explainability. Customization is required satisfying explainability desires within confidentiality and proportionality bounds. Findings advise technologists on prudent eXplainable AI technique selection given multi-dimensional tensions. Outcomes recommend policymakers balance worthy transparency goals with cohesive legislation enabling equitable dispute resolution. KEYWORDS K.4.1.c Ethics, K.4.1.g Regulation, Explainable AI, Trustworthy AI 1 https://joinup.ec.europa.eu/collection/ rolling-plan-ict-standardisation/rolling-plan-2022 2 https://eur-lex.europa.eu/legal-content/en/TXT/?uri= CELEX3A52021DC0118 IEEE Intelligent Systems

show abstract

Varieties of AI Explanations Under the Law. From the GDPR to the AIA, and Beyond

Cited by 19 publications

References 64 publications

The black box problem revisited. Real and imaginary challenges for automated legal decision making

The black box problem revisited. Real and imaginary challenges for automated legal decision making

Substantively smart cities – Participation, fundamental rights and temporality

Operationalizing Explainable Artificial Intelligence in the European Union Regulatory Ecosystem

Contact Info

Product

Resources

About