Zambrano, Guha & Henderson on Vulnerabilities in Discovery Tech

Diego Zambrano (Stanford), Neel Guha (Stanford), and Peter Henderson (Stanford) have posted “Vulnerabilities in Discovery Tech” (Harvard Journal of Law & Technology, (2022 Forthcoming) on SSRN. Here is the abstract:

Recent technological advances are changing the litigation landscape, especially in the context of discovery. For nearly two decades, technologies have reinvented document searches in complex litigation, normalizing the use of machine learning algorithms under the umbrella of “Technology Assisted Review” (TAR). But the latest technological developments are placing discovery beyond the reach of attorney understanding and firmly in the realm of computer science and engineering. As lawyers struggle to keep up, a creeping sense of anxiety is spreading in the legal profession about a lack of transparency and the potential for discovery abuse. Judges, attorneys, bar associations, and scholars warn that lawyers need to closely supervise the technical aspects of TAR and avoid the dangers of sabotage, intentional hacking, or abuse. But none of these commentators have defined with precision what the risks entail, furnished a clear outline of potential dangers, or defined the appropriate boundaries of debate.

This Article provides the first systematic assessment of the potential for abuse in technology-assisted discovery. The Article offers three contributions. First, our most basic aim is to provide a technical but accessible assessment of vulnerabilities in the TAR process. To do so, we use the latest computer science research to identify and catalogue the different ways that TAR can go awry, either due to intentional abuse or mistakes. Second, with a better understanding of how discovery can be subverted, we then map potential remedies and reassess current debates in a more helpful light. The upshot is that abuse of technology-assisted discovery is possible but can be preventable if the right review processes are in place. Finally, we propose reforms to improve the system in the short and medium term, with an emphasis on improved metrics that can more fully measure the quality of TAR. By exploring the technical background of discovery abuse, the Article demystifies the engineering substrate of modern discovery. Undertaking this study shows that lawyers can safeguard technology-assisted discovery without surrendering professional jurisdiction to engineers.

Sunstein on Governing by Algorithm? No Noise and (Potentially) Less Bias

Cass R. Sunstein (Harvard Law School) has posted “Governing by Algorithm? No Noise and (Potentially) Less Bias” on SSRN. Here is the abstract:

As intuitive statisticians, human beings suffer from identifiable biases, cognitive and otherwise. Human beings can also be “noisy,” in the sense that their judgments show unwanted variability. As a result, public institutions, including those that consist of administrative prosecutors and adjudicators, can be biased, noisy, or both. Both bias and noise produce errors. Algorithms eliminate noise, and that is important; to the extent that they do so, they prevent unequal treatment and reduce errors. In addition, algorithms do not use mental short-cuts; they rely on statistical predictors, which means that they can counteract or even eliminate cognitive biases. At the same time, the use of algorithms, by administrative agencies, raises many legitimate questions and doubts. Among other things, they can encode or perpetuate discrimination, perhaps because their inputs are based on discrimination, perhaps because what they are asked to predict is infected by discrimination. But if the goal is to eliminate discrimination, properly constructed algorithms nonetheless have a great deal of promise for administrative agencies.

Lessig on the First Amendment and Replicants

Lawrence Lessig (Harvard Law School) has posted “The First Amendment Does Not Protect Replicants” (Social Media and Democracy (Lee Bollinger & Geoffrey Stone, eds., Oxford 2022), Forthcoming) on SSRN. Here is the abstract:

As the semantic capability of computer systems increases, the law should resolve clearly whether the First Amendment protects machine speech. This essay argues it should not be read to reach sufficiently sophisticated — “replicant” — speech.

Rauch on Customized Speech and the First Amendment

Daniel Rauch (Yale Law School) has posted “Customized Speech and the First Amendment” (Harvard Journal of Law & Technology, Vol. 35, 2022 Forthcoming) on SSRN. Here is the abstract:

Customized Speech — speech targeted or tailored based on knowledge of one’s audience — is pervasive. It permeates our relationships, our culture, and, especially, our politics. Until recently, customization drew relatively little attention. Cambridge Analytica changed that. Since 2016, a consensus has decried Speech Customization as causing political manipulation, disunity, and destabilization. On this account, machine learning, social networks and Big Data make political Customized Speech a threat we constitutionally can, and normatively should, curtail.

That view is mistaken. In this Article, I offer the first systematic analysis of Customized Speech and the First Amendment. I reach two provocative results: Doctrinally, the First Amendment robustly protects Speech Customization. And normatively, even amidst Big Data, this protection can help society and democracy.

Doctrinally, the use of audience information to customize speech is, itself, core protected speech. Further, audience-information collection, while less protected, may still only be regulated by carefully drawn, content-neutral, generally applicable laws. And unless and until the state affirmatively enacts such laws (as, overwhelmingly, it has not), it may not curtail speakers’ otherwise-lawful use of such information in political Speech Customization.

What does this mean for democratic government? Today, Customized Speech raises fears about democratic discourse, hyper-partisan factions, and citizen autonomy. But these are less daunting than the consensus suggests, and are offset by key benefits: modern Customized Speech activates the apathetic, empowers the marginalized, and checks government overreach. Accordingly, many current proposals to restrict such Customized Speech — from disclosure requirements to outright bans — are neither constitutionally viable nor normatively required.