Search

The right to contest automated decisions as provided by Article 22 of the General Data Protection Regulation (GDPR) is a due process provision with concrete transparency implications. Based on this, the paper in hand aims, first, to provide an interpretation of Art 22 and the right to contest (as the key provision in determining the contours of transparency in relation to automated decisions under the GDPR); second, to provide a systematic account of possible administrative, procedural, and technical mechanisms (transparency measures) that could be deployed for the purpose contesting automated decisions; and third, to examine the compatibility of these mechanisms with the GDPR. Following the introduction, Part II starts with an analysis of the newly enacted right to contest solely automated decisions as provided under Article 22 of the GDPR. This part identifies the right to contest in Article 22 as the core remedy, with inherent transparency requirements which are foundational for due process. Setting the right to contest as the backbone of protection against the adverse effects of solely automated decisions, Part III focuses on certain key points and provisions under the GDPR, which are described as the 1st layer (human-intelligible) transparency. This part explores to what extent “information and access” rights (Articles 13, 14, and 15) could satisfy the transparency requirements for the purposes of contestation as explained in Part II. Next, Part IV briefly identifies the limits of 1st layer transparency – explaining how technical complexity together with competition and integrity-related concerns render human-level transparency either infeasible or legally impossible. In what follows, Part V conceptualizes a 2nd layer of transparency which consists of further administrative, procedural, and technical measures (i.e., design choices facilitating interpretability, institutional oversight, and algorithmic scrutiny). Finally, Part VI identifies four regulatory options, combining 1st and 2nd layer transparency measures to implement Article 22. The primary aim of the paper is to provide a systematic interpretation of Article 22 and examine how “the right to contest solely automated decisions” could help give meaning to the overall transparency provisions of the GDPR. With a view to transcend the current debates about the existence of a so-called right to an explanation, the paper develops an interdisciplinary approach, focusing on the specific transparency implications of the “right to contest” as a remedy of procedural nature.

This deliverable compiles the current EU legal and regulatory frameworks for data sharing and re-use in the transport sector (covering all transport modes) and identifies potential legal gaps that need to be addressed for the smooth operation of transport/mobility data spaces. The analysis provided in this document deals with both horizontally applicable and sector-specific (vertical) legislation.

Improvements in computer, communication and software technologies have stimulated systems that assume manpower functions online or independently. These “intelligent agents” are used for complex and large-scale information searches, data organization and electronic business transaction functions. These machines in question can be directly programmed to perform a particular function without any human intervention so that they can develop relevant reactions in accordance with the signals from the external world or communication networks. The most eminent feature of such software is to display interactions and purposive acts independent from the user. Some of their features can be listed as a) acting purposely without any direct instruction from the user; b) communicating with other sources of information; c) cooperating with other units or entities in order to attain a target result, d) adapting based on previous acts (the method of trial and error), e) reliability

Contains fulltext : 200776.pdf (Publisher’s version ) (Open Access)

Predictive policing has triggered a heated debate around the issue of ‘false positives’. Biased machine training can wrongly classify individuals as high risk simply as a result of belonging to a particular ethnic group and many agree such persons should not have to shoulder the burden of over-policing due to an inherent stochastic problem. The paper takes a pragmatic stand and argues that ‘true positives’, i.e. individuals who have been correctly identified as perpetrators, offer the best opportunity to address the issue of biased profiling. The first reason is purely pragmatic – they are already party to a criminal investigation and, as such, have a strong incentive to challenge law enforcement methods and scrutinize policing methods on an individual basis. The second reason is more general (and commonly subscribed to) – that discriminatory stops and searches are inherently unfair, threaten social peace, and frustrate targeted groups. To create an efficient legal tool against discriminatory law enforcement, defence should be entitled to contest a conviction for biased predictive policing, with a specific exclusionary rule protecting ‘true positives’ against the use of tainted evidence.

This book celebrates and mourns the increasing relevance of the 2008 volume of 'Profiling the European Citizen. Cross-Disciplinary Perspectives' (edited by Mireille Hildebrandt & Serge Gutwirth). Both volumes contain in-depth investigations by lawyers, philosophers and computer scientists into the legal, philosophical and computational background of the emerging algorithmic order. In BEING PROFILED:COGITAS ERGO SUM 23 scholars engage with the issues, underpinnings, operations and implications of micro-targeting, data-driven critical infrastructure, ethics-washing, p-hacking and democratic disruption. These issues have now become part of everyday life, reinforcing the urgency of the question: are we becoming what machines infer about us, or are we?, Paul Nemitz: Foreword: Profiling the European Citizen Mireille Hildebrandt: Introitus: what Descartes did not get Theories of normativity between law and machine learning Sylvie Delacroix: From agency-enhancement intentions to profile-based optimisation tools: what is lost in translation Patrick Allo: Mathematical values and the epistemology of data practices Seda Gürses et al.: Stirring the POTs: protective optimization technologies Emre Bayamlıoğlu: On the possibility of normative contestation of automated data-driven decisions Transparency theory for data-driven decision making Karen Yeung and Adrian Weller: How is ‘transparency’ understood by legal scholars and the machine learning community? Anton Vedder: Why data protection and transparency are not enough when facing social problems of machine learning in a big data context Jaap-Henk Hoepman: Transparency is the perfect cover-up (if the sun does not shine) Gloria González Fuster: Transparency as translation in data protection Presumption of innocence in data-driven government Lucia M. Sommerer: The presumption of innocence’s Janus head in data-driven government Sabine Gless: Predictive policing. In defence of ‘true positives’ Tobias Blanke: The geometric rationality of innocence in algorithmic decisions Linnet Taylor: On the presumption of innocence in data-driven government. Are we asking the right question? Legal and political theory in data-driven environments Orla Lynskey: A legal response to data-driven mergers Ben Wagner: Ethics as an escape from regulation. From “ethics-washing” to ethics-shopping? Arjen P. de Vries: Citizens in data land Saving machine learning from p-hacking Felix Stalder: From inter-subjectivity to multi-subjectivity: Knowledge claims and the digital condition Mireille Hildebrandt: Preregistration of machine learning research design. Against P-hacking Clare Ann Gollnick: Induction is not robust to search The legal and ML status of micro-targeting Bart Custers: Profiling as inferred data. Amplifier effects and positive feedback loops Liisa Janssens: A prospect of the future. How autonomous systems may qualify as legal persons Niels van Dijk: Profiles of personhood. On multiple arts of representing subjects Reuben Binns: Imagining data, between Laplace’s demon and the rule of succession

This brief provocation targets the mantra of the trade-off between accuracy and interpretability: the higher the accuracy, the lower the interpretability. It seems that this trade-off appeals to a deep-seated desire for magical thinking; the lure of things that work well even if we have no idea why. The suggestion is that in the realm of specific types of machine learning (ML), neither causality nor reasoning matters. Correlation and prediction are all that counts. The story goes that not just lay people, those using an ML application or those targeted by its decisions, but even those who developed the application cannot explain why it gets things right. I will confront this narrative from the perspective of ML research design, arguing that accuracy depends on the appropriate selection and curation of training and validation data, a properly articulated machine-readable task, a well-developed hypotheses space, and the selection of a relevant performance metric. The latter, indeed, may give rise to P-hacking: cherry picking the most favourable performance metric. This entails that accuracy in the realm of data should not be conflated with correctness in the realm of atoms. In other words, if we cannot explain why an ML application gets things right, we cannot be sure that it gets things right.

Contains fulltext : 200790.pdf (Publisher’s version ) (Open Access)

Contains fulltext : 236275.pdf (Publisher’s version ) (Open Access)

ispartof: Being profiled: Cogitas ergo sum pages:42-45 ispartof: pages:42-45 status: published