Judgement-Proof Robots and Artificial Intelligence A Comparative Law and Economics Approach Mitja Kovač Judgement-Proof Robots and Artificial Intelligence Mitja Kovacˇ Judgement-Proof Robots and Artificial Intelligence A Comparative Law and Economics Approach Mitja Kovacˇ School of Economics and Business University of Ljubljana Ljubljana, Slovenia ISBN 978-3-030-53643-5 ISBN 978-3-030-53644-2 (eBook) https://doi.org/10.1007/978-3-030-53644-2 © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2020 This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval,electronicadaptation,computersoftware,orbysimilarordissimilarmethodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such namesareexemptfromtherelevantprotectivelawsandregulationsandthereforefreefor general use. Thepublisher,theauthorsandtheeditorsaresafetoassumethattheadviceandinforma- tion in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respecttothematerialcontainedhereinorforanyerrorsoromissionsthatmayhavebeen made.Thepublisherremainsneutralwithregardtojurisdictionalclaimsinpublishedmaps and institutional affiliations. Cover credit: © John Rawsterne/patternhead.com This Palgrave Macmillan imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland Preface Thisbookisaboutthefutureofunimaginableprogresswherethewildest dreamsofvisionaryscientistsmaterialized.Itisaboutafuturisticworldin whichsuper-intelligent,superhumanartificialintelligenceservesashuman compatible and, among other things, acts in its own will. It is also about unprecedented and currently uncontemplated hazards that such super- human and super-intelligent artificial intelligence may impose on human societies. However, this book is not about a super-intelligent AI that is conscious, since no one working in the AI field is attempting to make machines conscious. Famous Hollywood movies like “I robot” where a god detective Spooner—alias Will Smith—chases hordes of evil and conscious robots attempting to enslave humans are actually missing the point. It is competence, and not consciousness, that matters. Namely, if onewritesanalgorithmthatwhenrunningwillformandcarryoutaplan which will result in significant damages to life or property, unforeseeable hazards or even in the destruction of a human race, then it is not about the AI’s consciousness but about its competence and capacity. Of course, no one can predict exactly how the AI will develop but undoubtedly it will be the dominant technology of the future. Neverthe- less, policymakers and lawmakers must prepare ex ante for the possibility that AI will become super-intelligent and that its actions might cause severe damages and hazards. This book is thus an attempt to provide a law and economics treatment of such uncontemplated development and shouldberegardedasacontributiontolawmakersandlegalpractitioners v vi PREFACE around the world to learn how to avoid the risks, mitigate potential hazards and to offer a set of regulatory tools that could be employed in ex ante controlling, regulating the potentially biggest event in human history. This book could not have been made without the enthusiastic support of my parents and my loved ones. MyspecialgratitudegoesalsotoProfessorGerritDeGeestfromWash- ington University in St. Louis for his extraordinary, fascinating, inspiring, and path-breaking discussions and lectures which I have been privileged to follow and admire. I am especially indebted to and would like to express our sincere gratitude for their precious substantive comments, insights, reflections, feedbacks, suggestions, discussions, and inspiration to: Paul Aubrecht, Nick van der Beek, Roger van den Bergh, John Bell, Greta Bosch, Boudewijn Bouckaert, Marianne Breier, Miriam Buiten, Giuseppe Dari- Mattiachi, Gerrit De Geest, Ben Depoorter, Larry DiMatteo, Thomas Eger, Jan Essink, Michael Faure, Luigi Franzoni, Nuno Garupa, Paula Giliker, Victor Goldberg, James Gordley, Alice Guerra, Eric Helland, JohandenHertogSvenHoeppner,RolandKirstein,JonathanKlick,Anne Lafarre, Henrik Lando, Igor Loncarski, Anthony Ogus, Vernon Palmer, Francesco Parisi, Alessio Pacces, Catherine Pedamon, Roy Pertain, Christina Poncibo, Jens Prüffer, Elena Reznichenko, Wolf-Georg Ringe, Hans-Bernd Schäfer, Matej Marinc, Marcus Smith, Dusan Mramor, Nancy Van Nuffel, Holger Spamann, Rok Spruk, Christoph Van der Elst, Ann-Sophie Vandenberghe, Stefan Voight, Franziska Weber, Wicher Schreuders, Louis Visscher, Spela Vizjak, Elisabeth Wielinger, and Wolf- gang Weigel. I am also grateful to Miha Škerlevaj, Sandra Duraševic´, Martina Petan, IvanaPranjic´,DunjaZlotrg,ErnaEmric´,TadejaŽabkar,RebekaKoncilja, andVesnaŽabkarfortheirdaily,round-a-clockcareandimmenseorgani- zational support. This is also the place to thank the publisher Palgrave Macmillan on behalf of all contributing authors in particular to Ruth Jenner,ArunKumarandRuthNobleastheresponsiblepublisherofficers. IcouldnothavecompletedthisbookwithoutthesupportofSlovenian ResearchAgency(AgencijazaraziskovalnodejavnostRepublikeSlovenije), since this book is part of our project “Challenges of inclusive sustain- abledevelopmentinthepredominantparadigmofeconomicandbusiness sciences” (P5-0128). PREFACE vii Finally, thanks are due to my Dean Professor Metka Tekavcˇicˇ (School of Economics and Business University of Ljubljana) and to all of my colleagues from the University of Ljubljana. This book has been written in the times of stress when the Covid- 19 pandemic locked down the entire European continent. The project itself has been, during my teaching visits at the Erasmus University of RotterdamandattheGhentUniversity,graduallydevelopedoverthepast three years. I do hope that you will enjoy it. Ljubljana, Slovenia Mitja Kovacˇ Ghent, Belgium Rotterdam, The Netherlands May 2020 Contents 1 Introduction 1 Bibliography 9 Part I Conceptual Framework 2 Economic Analysis of Law 13 1 Introduction 13 2 On the Nature of Economic Reasoning 15 3 Methodology and Concepts Used in the Economic Analysis of Law 16 4 Comparative Law and Economics 20 5 Behavioural Law and Economics 22 6 Obstacles to an Economic Approach 26 7 Conclusions 27 Bibliography 27 3 The Case for Regulatory Intervention and Its Limits 33 1 Introduction 33 2 A Nirvana World: Perfect Competition 35 3 Market Failures 36 4 Nature, Scope, and Form of Regulation 40 5 Conclusion 42 Bibliography 42 ix x CONTENTS 4 Introduction to the Autonomous Artificial Intelligence Systems 47 1 Introduction 47 2 A General Background and Key Concepts 48 3 Setting the Scene: Definitions, Concepts, and Research Trends 51 4 Learning and Communicating 56 5 Robotics 57 6 Conclusion 59 Bibliography 59 Part II Judgement-Proof Superintelligent and Superhuman AI 5 What Can Get Wrong? 67 1 Introduction 67 2 Can AI Think and Act Intelligently? 69 3 Risks of Developing Artificial Intelligence 71 4 AI Making Moral Choices and Independent Development 74 5 Conclusion 75 Bibliography 76 6 Judgement-proof Problem and Superhuman AI Agents 79 1 Introduction 80 2 Low of Torts: Responsibility and Liability 82 3 Tort Law and Economics 88 4 Legal Concept of Agency and Superhuman AI 90 5 Causation and Superhuman Artificial Intelligence 91 6 Judgement-proof Problem 94 7 Judgement-proof Superhuman Artificial Intelligence 97 8 Conclusion 101 Bibliography 102 7 Towards Optimal Regulatory Framework: Ex Ante Regulation of Risks and Hazards 109 1 Introduction 109 CONTENTS xi 2 How to Deal with Judgement-Proof Super-Intelligent AI Agents 112 3 Special Electronic Legal Personality 122 4 Tinbergen Golden Rule of Thumb and Optimal Regulatory Timing 124 5 Liability for Harm Versus Safety Regulation 127 6 Regulatory Sandboxes 128 7 Liability for Harm and Incentives to Innovate 131 8 Historical Legal Responses to Technical Innovations: Anti-fragile Law 132 9 Current Trends in Legislative Activity 137 10 Conclusions 140 Bibliography 140 Epilogue 145 Index 149