Coglianese & Lai on Assessing Automated Administration

Cary Coglianese (University of Pennsylvania Carey Law School) and Alicia Lai (same) have posted “Assessing Automated Administration” (In Oxford Handbook on AI Governance (Justin Bullock et al. eds., forthcoming)). Here is the abstract:

To fulfill their responsibilities, governments rely on administrators and employees who, simply because they are human, are prone to individual and group decision-making errors. These errors have at times produced both major tragedies and minor inefficiencies. One potential strategy for overcoming cognitive limitations and group fallibilities is to invest in artificial intelligence (AI) tools that allow for the automation of governmental tasks, thereby reducing reliance on human decision-making. Yet as much as AI tools show promise for improving public administration, automation itself can fail or can generate controversy. Public administrators face the question of when exactly they should use automation. This paper considers the justifications for governmental reliance on AI along with the legal concerns raised by such reliance. Comparing AI-driven automation with a status quo that relies on human decision-making, the paper provides public administrators with guidance for making decisions about AI use. After explaining why prevailing legal doctrines present no intrinsic obstacle to governmental use of AI, the paper presents considerations for administrators to use in choosing when and how to automate existing processes. It recommends that administrators ask whether their contemplated uses meet the preconditions for the deployment of AI tools and whether these tools are in fact likely to outperform the status quo. In moving forward, administrators should also consider the possibility that a contemplated AI use will generate public or legal controversy, and then plan accordingly. The promise and legality of automated administration ultimately depends on making responsible decisions about when and how to deploy this technology.

Ashley on Capturing the Dialectic between Principles and Cases

Kevin Ashley (University of Pittsburgh – School of Law) has posted “Capturing the Dialectic between Principles and Cases” (Jurimetrics, Vol. 44, p. 229, 2004) on SSRN. Here is the abstract:

Theorists in ethics and law posit a dialectical relationship between principles and cases; abstract principles both inform and are informed by the decisions of specific cases. Until recently, however, it has not been possible to investigate or confirm this relationship empirically. This work involves a systematic study of a set of ethics cases written by a professional association’s board of ethical review. Like judges, the board explains its decisions in opinions. It applies normative standards, namely principles from a code of ethics, and cites past cases. We hypothesized that the board’s explanations of its decisions elaborated upon the meaning and applicability of the abstract code principles and past cases. In effect, the board operationalizes the principles and cases. We hypothesized further that this operationalization could be captured computationally and used to improve automated information retrieval. A computer program was designed to retrieve from the on-line database those ethics code principles and past cases that are relevant to analyzing new problems. In an experiment, we used the computer program to test the hypotheses. The experiment demonstrated that the dialectical relationship between principles and cases exists and that the associated operationalization information improves the program’s ability to assess which codes and cases are relevant to analyzing new problems. The results have significance both to the study of legal reasoning and improvement of legal information retrieval.

Malgieri & Pasquale on Ex Ante Accountability for AI

Gianclaudio Malgieri (EDHEC; Vrije Universiteit Brussel Law) and Frank A. Pasquale (Brooklyn Law School) have posted “From Transparency to Justification: Toward Ex Ante Accountability for AI” on SSRN. Here is the abstract:

At present, policymakers tend to presume that AI used by firms is legal, and only investigate and regulate when there is suspicion of wrongdoing. What if the presumption were flipped? That is, what if a firm had to demonstrate that its AI met clear requirements for security, non-discrimination, accuracy, appropriateness, and correctability, before it was deployed? This paper proposes a system of “unlawfulness by default” for AI systems, an ex-ante model where some AI developers have the burden of proof to demonstrate that their technology is not discriminatory, not manipulative, not unfair, not inaccurate, and not illegitimate in its legal bases and purposes. The EU’s GDPR and proposed AI Act tend toward a sustainable environment of AI systems. However, they are still too lenient and the sanction in case of non-conformity with the Regulation is a monetary sanction, not a prohibition. This paper proposes a pre-approval model in which some AI developers, before launching their systems into the market, must perform a preliminary risk assessment of their technology followed by a self-certification. If the risk assessment proves that these systems are at high-risk, an approval request (to a strict regulatory authority, like a Data Protection Agency) should follow. In other terms, we propose a presumption of unlawfulness for high-risk models, while the AI developers should have the burden of proof to justify why the AI is not illegitimate (and thus not unfair, not discriminatory, and not inaccurate). Such a standard may not seem administrable now, given the widespread and rapid use of AI at firms of all sizes. But such requirements could be applied, at first, to the largest firms’ most troubling practices, and only gradually (if at all) to smaller firms and less menacing practices.

Shope on NGO Engagement in the Age of Artificial Intelligence

Mark Shope (National Yang Ming Chiao Tung University; Indiana University Robert H. McKinney School of Law) has posted “NGO Engagement in the Age of Artificial Intelligence” (Buffalo Human Rights Law Review, Vol. 28, pp. 119-158, 2022) on SSRN. Here is the abstract:

From AI and human rights focused NGOs to thematic NGOs whose subjects are impacted by AI, the AI and human rights discourse within NGOs has moved from simply keeping an eye on AI to being an integral part of NGO work. At the same time, the issue of AI and human rights is being addressed by governments in their policymaking and rulemaking to, for example, protect human rights and remain compliant with their responsibilities under international human rights instruments. When governments are reporting to United Nations treaty bodies as required under international human rights instruments, and the reports and communications include topics of artificial intelligence, how and to what extent are NGOs engaging in this dialogue? This article explores how artificial intelligence can impact rights under the nine core human rights instruments and how NGOs should monitor States parties under these instruments, providing suggestions to guide NGO engagement in the reporting process.

Li on Affinity-Based Algorithmic Pricing: A Dilemma for EU Data Protection Law

Zihao Li (University of Glasgow) has posted “Affinity-Based Algorithmic Pricing: A Dilemma for EU Data Protection Law” (Computer Law & Security Review, Volume 46, 2022) on SSRN. Here is the abstract:

The emergence of big data and machine learning has allowed sellers and online platforms to tailor pricing for customers in real-time, but as many legal scholars have pointed out, personalised pricing poses a threat to the fundamental values of privacy and non-discrimination, raising legal and ethical concerns. However, most of those studies neglect affinity-based algorithmic pricing, which may bypass the General Data Protection Regulation (GDPR). This paper evaluates current data protection law in Europe against online algorithmic pricing. The first contribution of the paper is to introduce and clarify the term “online algorithmic pricing” in the context of data protection legal studies, as well as a new taxonomy of online algorithmic pricing by processing the data types. In doing so, the paper finds that the legal nature of affinity data is hard to classify as personal data. Therefore, affinity-based algorithmic pricing is highly likely to circumvent the GDPR. The second contribution of the paper is that it points out that even though some types of online algorithmic pricing can be covered by the GDPR, the data rights provided by the GDPR struggle to provide substantial help. The key finding of this paper is that the GDPR fails to apply to affinity-based algorithmic pricing, but the latter still can lead to privacy invasion. Therefore, four potential resolutions are raised, relating to group privacy, the remit of data protection law, the ex-ante measures in data protection, and a more comprehensive regulatory approach.

Schrepel & Goroza on The Adoption of Computational Antitrust by Agencies: 2021 Report

Thibault Schrepel (University Paris 1 Panthéon-Sorbonne; VU University Amsterdam; Stanford University’s Codex Center; Sciences Po) and Teodora Groza (Sciences Po Law School) have posted “The Adoption of Computational Antitrust by Agencies: 2021 Report” (2 Stanford Computational Antitrust, 78 (2022)) on SSRN. Here is the abstract:

In the first quarter of 2022, the Stanford Computational Antitrust project team invited the partnering antitrust agencies to share their advances in implementing computational tools. Here are the results of the survey.