Three Laws

Sarel on Restraining ChatGPT @dr_roee @unihh

Roee Sarel, Institute of Law and Economics, University of Hamburg, is publishing Restraining ChatGPT in the UC Law SF Journal. Here is the abstract.

ChatGPT is a prominent example of how Artificial Intelligence (AI) has stormed into our lives. Within a matter of weeks, this new AI—which produces coherent and human-like textual answers to questions—has managed to become an object of both admiration and anxiety. Can we trust generative AI systems, such as ChatGPT, without regulatory oversight? Designing an effective legal framework for AI requires answering three main questions: (i) is there a market failure that requires legal intervention? (ii) should AI be governed through public regulation, tort liability, or a mixture of both? and (iii) should liability be based on strict liability or a fault-based regime such as negligence? The law and economics literature offers clear considerations for these choices, focusing on the incentives of injurers and victims to take precautions, engage in efficient activity levels, and acquire information. This Article is the first to comprehensively apply these considerations to ChatGPT as a leading test case. As the United States is lagging behind in its response to the AI revolution, I focus on the recent proposals in the European Union to restrain AI systems, which apply a risk-based approach and combine regulation and liability. The analysis reveals that this approach does not map neatly onto the relevant distinctions in law and economics, such as market failures, unilateral versus bilateral care, and known versus unknown risks. Hence, the existing proposals may lead to various incentive distortions and inefficiencies. The Article, therefore, calls upon regulators to place a stronger emphasis on law and economics concepts in their design of AI policy.

Download the article from SSRN at the link.

Schwalbe on Algorithms, Machine Learning, and Collusion

Ulrich Schwalbe, University of Hohenheim, has published Algorithms, Machine Learning, and Collusion. Here is the abstract.

This paper discusses the question whether self-learning price-setting algorithms are able to coordinate their pricing behaviour to achieve a collusive outcome that maximizes the joint profits of the firms using these algorithms. While the legal literature generally assumes that algorithmic collusion is indeed possible and in fact very easy, the computer science literature on cooperation between algorithms as well as the economics literature on collusion in experimental oligopolies indicate that a coordinated and in particular tacitly collusive behaviour is in general rather difficult to achieve. Many studies have shown that some form of communication is of vital importance for collusion if there are more than two firms in a market. Communication between algorithms is also a topic in artificial intelligence research and some recent contributions indicate that algorithms may learn to communicate, albeit in a rather limited way. This leads to the conclusion that algorithmic collusion is currently much more difficult to achieve than often assumed in the legal literature and is therefore currently not a particularly important competitive concern. In addition, there are also several legal problems associated with algorithmic collusion, for example, questions of liability, of auditing and monitoring algorithms as well as enforcement. The limited resources of competition authorities should rather be devoted to more pressing problems as, for example, the abuse of dominant positions by large online-platforms.

Download the article from SSRN at the link.

Ringe and Ruof on a Regulatory Sandbox for Robo Advice

Wolf-Georg Ringe and Christopher Ruof, both of the University of Hamburg, Institute of Law & Economics, have published A Regulatory Sandbox for Robo Advice as European Banking Institute Working Paper Series 2018 - no. 26. Here is the abstract.

Robo advice, the automated provision of financial advice without human intervention, holds the promise of cheap, convenient and fast investment services for consumers – freed from human error or bias. However, retail investors have limited capacity to assess the soundness of the advice, and are prone to make hasty, unverified investment decisions. Moreover, financial advice based on rough and broad classifications may fail to take into account the individual preferences and needs of the investor. On a more general scale, robo advice may be a source of new systemic risk. At this stage, the existing EU regulatory framework is of little help. Instead, this paper proposes a regulatory “sandbox” – an experimentation space – as a step towards a regulatory environment where such new business models can thrive. A sandbox would allow market participants to test robo advice services in the real market, with real consumers, but under close scrutiny of the supervisor. The benefit of such an approach is that it fuels the development of new business practices and reduces the “time to market” cycle of financial innovation while simultaneously safeguarding consumer protection. At the same time, a sandbox allows for mutual learning in a field concerning a little-known phenomenon, both for firms and for the regulator. This would help reducing the prevalent regulatory uncertainty for all market participants. In the particular EU legal framework with various layers of legal instruments, the implementation of such a sandbox is not straightforward. In this paper, we propose a “guided sandbox”, operated by the EU Member States, but with endorsement, support, and monitoring by EU institutions. This innovative approach would be somewhat unchartered territory for the EU, and thereby also contribute to the future development of EU financial market governance.

Download the article from SSRN at the link.

Segal on Legal Jobs in the Age of AI: Moving From Today's Limited Universe of Data Toward the Great Beyond @CGIntelligence @SavLawRev

Philip Segal, Charles Griffin Intelligence, LLC, is publishing Legal Jobs in the Age of Artificial Intelligence: Moving from Today's Limited Universe of Data Toward the Great Beyond in the Savannah Law Review (2018). Here is the abstract.

Lawyers are unprepared for the major differences between the closed or highly limited universes of data with which today’s most popular AI programs deal, and the vastly larger, unstructured field of data that the AI of tomorrow will attempt to conquer. In the space of unlimited data – at least on the fact gathering side – this article argues that there are dozens of AI applications currently in development that could help lawyers do their jobs better. The impact on legal employment from these unlimited-fact universe programs does not appear to be negative, for the simple reason that they represent a chance to look at data previously unexamined by lawyers. But, to be useful, these programs will demand skills and a job outlook not always in evidence among lawyers. Even today, software used in law firms is not employed to its full potential. Lawyers often miss the point that while computers work purely in a deductive way, higher human functioning is often not deductive at all. What kind of person does a firm want to have running all the sophisticated machinery? In brief, it may not be the ideal law student of today. The paper concludes with suggestions about how to train lawyers who will be using AI to use it wisely. Those expecting to type in a few queries and get a final correct answer will be sorely disappointed.

Download the article from SSRN at the link.

Petit on Antitrust and Artificial Intelligence: A Research Agenda @CompetitionProf

Nicolas Petit, University of Liege School of Law; International Center for Law & Economics; University of South Australia School of Law, has published Antitrust and Artificial Intelligence - A Research Agenda. Here is the abstract.

This short paper provides an overview of the claims made in the emerging scholarship on artificial intelligence and antitrust policy. It discusses in particular the conjecture that markets will be rife with algorithmic collusion and extractive personalized pricing. It stresses the main methodological, theoretical and empirical limitations that underpin the claims made in the scholarship, and concludes with a call for evidenced-based antitrust policy.

Download the article from SSRN at the link.

Petit on Antitrust and AI: A Research Agenda @CompetitionProf

Nicolas Petit, University of Liege, School of Law; International Center for Law & Economics; University of South Australia School of Law, hs published Antitrust and Artificial Intelligence - A Research Agenda. Here is the abstract.

This short paper provides an overview of the claims made in the emerging scholarship on artificial intelligence and antitrust policy. It discusses in particular the conjecture that markets will be rife with algorithmic collusion and extractive personalized pricing. It stresses the main methodological, theoretical and empirical limitations that underpin the claims made in the scholarship, and concludes with a call for evidenced-based antitrust policy.

Download the paper from SSRN at the link.