Man-cho So on Technical Elements of Machine Learning for Intellectual Property Law

Anthony Man-cho So (The Chinese University of Hong Kong (CUHK)) has posted “Technical Elements of Machine Learning for Intellectual Property Law” (Artificial Intelligence and Intellectual Property, 2020) on SSRN. Here is the abstract:

Recent advances in artificial intelligence (AI) technologies have transformed our lives in profound ways. Indeed, AI has not only enabled machines to see (e.g., face recognition), hear (e.g., music retrieval), speak (e.g., speech synthesis), and read (e.g., text processing), but also, so it seems, given machines the ability to think (e.g., board game-playing) and create (e.g., artwork generation). This chapter introduces the key technical elements of machine learning (ML), which is a rapidly growing sub-field in AI and drives many of the aforementioned applications. The goal is to elucidate the ways human efforts are involved in the development of ML solutions, so as to facilitate legal discussions on intellectual property issues.

Abbot on the Reasonable Robot

Ryan Abbot (University of Surrey School of Law, University of California, Los Angeles – David Geffen School of Medicine) has posted an excerpt from his book “The Reasonable Robot: Artificial Intelligence and the Law” on SSRN. Here is the abstract:

AI and people do not compete on a level-playing field. Self-driving vehicles may be safer than human drivers, but laws often penalize such technology. People may provide superior customer service, but businesses are automating to reduce their taxes. AI may innovate more effectively, but an antiquated legal framework constrains inventive AI. In The Reasonable Robot, Ryan Abbott argues that the law should not discriminate between AI and human behavior and proposes a new legal principle that will ultimately improve human well-being. This work should be read by anyone interested in the rapidly evolving relationship between AI and the law.

Armijo on Reasonableness as Censorship: Section 230 Reform, Content Moderation, and The First Amendment

Enrique Armijo (Elon University School of Law) has posted “Reasonableness as Censorship: Section 230 Reform, Content Moderation, and The First Amendment” (Florida Law Review, Forthcoming) on SSRN. Here is the abstract:

For the first time in the Internet’s history, revising the Communications Decency Act’s Section 230 to permit greater liability for social media platforms’ carriage of third-party content seems to many not just viable, but necessary. Most of these calls are built around the longstanding common law liability principles of duty and reasonableness. The use of reasonableness in the Section 230 context would condition the liability of social media platforms on a requirement that the platforms “take reasonable steps to prevent or address unlawful uses of their services.” These reforms are finding common cause with several legislative and executive efforts seeking to compel platforms to adhere to “reasonable” or “politically neutral” moderation policies, or else face increased liability. And calls for entirely new regulatory regimes for social media, some of which call for new federal agencies to implement them, advocate for similar approaches.

This Article is the first comprehensive response to these efforts. Using the guidance of the common law to unpack the connections between reasonableness, imminence, and intermediary liability, the Article argues that these proposed reforms are misguided as a matter of technology and information policy and are so legally dubious that they have little chance of surviving the court challenges that would inevitably follow their adoption. It demonstrates the many problems associated with adopting a common law-derived standard of civil liability like “reasonableness” as a regulatory baseline for prospective platform intermediary fault. The Article also discusses the challenges that the use of Artificial Intelligence-driven content moderation presents to the task of defining reasonableness, and considers the fit between content moderation and products liability, another common law fault theory increasingly used to argue expanding intermediary liability’s scope. “Reasonableness”-based Section 230 reforms would also lead to unintended, speech-averse results. And even if Section 230 were to be legislatively revised, serious constitutional problems would remain with respect to holding social media platforms liable, either civilly or criminally, for third-party user content.

Ong & Loo on Gauging the Acceptance of Contact Tracing Technology

Ee-Ing Ong (Singapore Management University School of Law, Singapore Management University – Centre for AI & Data Governance) and Wee Ling Loo (Singapore Management University School of Law, Singapore Management University – Centre for AI & Data Governance) have posted “Gauging the Acceptance of Contact Tracing Technology: An Empirical Study of Singapore Residents’ Concerns and Trust in Information Sharing” (Regulatory Insights on Artificial Intelligence: Research for Policy 2021) on SSRN. Here is the abstract:

In response to the COVID-19 pandemic, governments began implementing various forms of contact tracing technology. Singapore’s implementation of its contact tracing technology, TraceTogether, however, was met with significant concern by its population, with regard to privacy and data security. This concern did not fit with the general perception that Singaporeans have a high level of trust in its government. We explore this disconnect, using responses to our survey (conducted pre-COVID-19) in which we asked participants about their level of concern with the government and business collecting certain categories of personal data. The results show that respondents had less concern with the government as compared to a business collecting most forms of personal data. Nonetheless, they still had a moderately high level of concern about sharing such data with the government. We further found that income, education and perceived self-exposure to AI are associated with higher levels of concern with the government collecting personal data relevant to contact tracing, namely health history, location and social network friends’ information. This has implications for Singapore residents’ trust in government collecting data and hence the success of such projects, not just for contact tracing purposes but for other government-related data collection undertakings.

Seah on “Nose to Glass: Looking In to Get Beyond”

Joseph Seah (Singapore Management University) has posted “Nose to Glass: Looking In to Get Beyond” (Navigating the Broader Impacts of AI Research Workshop at NeurIPS 2020) on SSRN. Here is the abstract:

Brought into the public discourse through investigative work by journalists and scholars, awareness of algorithmic harms is at an all-time high. An increasing amount of research has been dedicated to enhancing responsible artificial intelligence (AI), with the goal of addressing, alleviating, and eventually mitigating the harms brought on by the roll out of algorithmic systems. Nonetheless, implementation of such tools remains low. Given this gap, this paper offers a modest proposal: that the field–particularly researchers concerned with responsible research and innovation–may stand to gain from supporting and prioritising more ethnographic work. This embedded work can flesh out implementation frictions and reveal organisational and institutional norms that existing work on responsible artificial intelligence has not yet been able to offer. In turn, this can contribute to more insights about the anticipation of risks and mitigation of harm. This paper reviews similar empirical work typically found elsewhere–commonly in science and technology studies and safety science research–and lays out challenges of this form of inquiry.

Recommended.

Chen on Interpreting Linear Beta Coefficients Alongside Feature Importance in Machine Learning

James Ming Chen (Michigan State University – College of Law) has posted “Linear Beta Coefficients Alongside Feature Importance in Machine Learning” on SSRN. Here is the abstract:

Machine-learning regression models lack the interpretability of their conventional linear counterparts. Tree- and forest-based models offer feature importances, a vector of probabilities indicating the impact of each predictive variable on a model’s results. This brief note describes how to interpret the beta coefficients of the corresponding linear model so that they may be compared directly to feature importances in machine learning.

Wachter, Mittelstadt & Russel on Bias Preservation in Machine Learning: The Legality of Fairness Metrics Under EU Non-Discrimination Law

Sandra Wachter (University of Oxford – Oxford Internet Institute), Brent Mittelstadt (University of Oxford – Oxford Internet Institute), and Chris Russell (Amazon Web Services, Inc.) have posted “Bias Preservation in Machine Learning: The Legality of Fairness Metrics Under EU Non-Discrimination Law” (West Virginia Law Review, Forthcoming) on SSRN. Here is the abstract:

Western societies are marked by diverse and extensive biases and inequality that are unavoidably embedded in the data used to train machine learning. Algorithms trained on biased data will, without intervention, produce biased outcomes and increase the inequality experienced by historically disadvantaged groups. Recognising this problem, much work has emerged in recent years to test for bias in machine learning and AI systems using various fairness and bias metrics. Often these metrics address technical bias but ignore the underlying causes of inequality. In this paper we make three contributions. First, we assess the compatibility of fairness metrics used in machine learning against the aims and purpose of EU non-discrimination law. We show that the fundamental aim of the law is not only to prevent ongoing discrimination, but also to change society, policies, and practices to ‘level the playing field’ and achieve substantive rather than merely formal equality. Based on this, we then propose a novel classification scheme for fairness metrics in machine learning based on how they handle pre-existing bias and thus align with the aims of non-discrimination law. Specifically, we distinguish between ‘bias preserving’ and ‘bias transforming’ fairness metrics. Our classification system is intended to bridge the gap between non-discrimination law and decisions around how to measure fairness in machine learning and AI in practice. Finally, we show that the legal need for justification in cases of indirect discrimination can impose additional obligations on developers, deployers, and users that choose to use bias preserving fairness metrics when making decisions about individuals because they can give rise to prima facie discrimination. To achieve substantive equality in practice, and thus meet the aims of the law, we instead recommend using bias transforming metrics. To conclude, we provide concrete recommendations including a user-friendly checklist for choosing the most appropriate fairness metric for uses of machine learning and AI under EU non-discrimination law.

Rozenshtein on Cost-Benefit Analysis and the Digital Fourth Amendment

Alan Z. Rozenshtein (University of Minnesota Law School) has posted “Cost-Benefit Analysis and the Digital Fourth Amendment” (40 Criminal Justice Ethics (2021 Forthcoming)) (reviewing Review of Ric Simmons, Smart Surveillance: How to Interpret the Fourth Amendment in the Twenty-First Century (2019)) on SSRN. Here is the abstract:

In “Smart Surveillance,” Ric Simmons argues for the application of cost-benefit analysis (CBA) to digital surveillance. This review argues that, although Simmons is right to look to CBA as a tool for applying the Fourth Amendment to new technology, his faith in the courts as the main practitioners of surveillance CBA is misguided. Across a variety of dimensions of institutional competence, the political branches, not the courts, are best placed to make surveillance policy under conditions of technological change.

Richardson on Defining and Demystifying Automated Decision Systems

Rashida Richardson (Rutgers, The State University of New Jersey – Rutgers Law School, Northeastern University School of Law) has posted “Defining and Demystifying Automated Decision Systems” (Maryland Law Review, Forthcoming) on SSRN. Here is the abstract:

Government agencies are increasingly using automated decision systems to aid or supplant human decision-making and policy enforcement in various sensitive social domains. They determine who will have their food subsidies terminated, how much healthcare benefits a person is entitled to, and who is likely to be a victim of a crime. Yet, existing legislative and regulatory definitions fail to adequately describe or clarify how these technologies are used in practice and their impact on society. This failure to adequately describe and define “automated decision systems” leads to such systems evading scrutiny that policymakers are increasingly recognizing is warranted and potentially impedes avenues for legal redress. Such oversights can have concrete consequences for individuals and communities, such as increased law enforcement harassment, deportation, denial of housing or employment opportunities, and death.

This article is the first in law review literature to provide two clear and measured definitions of “automated decision systems” for legislative and regulatory purposes and to suggest how these definitions should be applied. The definitions and analytical framework offered in this article clarify automated decision systems as prominent modes of governance and social control that warrant greater public scrutiny and immediate regulation. The definitions foreground the social implications of these technologies in addition to capturing the multifarious functions these technologies perform as they relate to rights, liberties, public safety, access, and opportunities. To demonstrate the significance and practicality of these definitions I analyze and apply them to two modern use cases: teacher evaluation systems and gang databases. I then explore how policymakers should determine exemptions and evaluate two technologies routinely used in government: email filters and accounting software. This law review provides a much-needed intervention in global public policy discourse and interdisciplinary scholarship regarding the regulation of emergent, data- driven technologies.

Narechania on Machine Learning as Natural Monopoly

Tejas N. Narechania (University of California, Berkeley, School of Law) has posted “Machine Learning as Natural Monopoly” (Iowa Law Review, Forthcoming) on SSRN. Here is the abstract:

Machine learning is transforming the economy, reshaping operations in communications, law enforcement, and medicine, among other sectors. But all is not well: It is now well-established that many machine-learning-based applications harvest vast amounts of personal information and yield results that are systematically biased. In response, policymakers have begun to offer a range of inchoate and often insufficient solutions, overlooking the possibility—suggested intuitively by scholars across disciplines—that these systems are natural monopolies, and thus neglecting the long legal tradition of natural monopoly regulation.

Drawing on the computer science, economics, and legal literatures, I find that machine-learning-based applications can be natural monopolies. Several features of machine learning suggest that this is so, including the fixed costs of developing these applications and the computational methods of optimizing these systems. This conclusion yields concrete policy implications: Where natural monopolies exist, public oversight and regulation is typically superior to market discipline through competition. Hence, where machine-learning-based applications are natural monopolies, this regulatory tradition offers one framework for confronting a range of issues—from privacy to accuracy and bias—that attend to such systems. Just as prior natural monopolies—the railways, electric grids, and telephone networks—faced rate and service regulation to protect against extractive, anticompetitive, and undemocratic behaviors, so too might machine-learning-based applications face similar public regulation to limit intrusive data collection and protect against algorithmic redlining, among other harms.