13203 research outputs found
Sort by
Vicarious Liability for AI
Get PDFWhen an algorithm harms someone—say by discriminating against her, exposing her personal data, or buying her stock using inside information—who should pay? If that harm is criminal, who deserves punishment? In ordinary cases, when A harms B, the first step in the liability analysis turns on what sort of thing A is. If A is a natural phenomenon, like a typhoon or mudslide, B pays, and no one is punished. If A is a person, then A might be liable for damages and sanction. The trouble with algorithms is that neither paradigm fits. Algorithms are trainable artifacts with “off” switches, not natural phenomena. They are not people either, as a matter of law or metaphysics.
An appealing way out of this dilemma would start by complicating the standard A-harms-B scenario. It would recognize that a third party, C, usually lurks nearby when an algorithm causes harm, and that third party is a person (legal or natural). By holding third parties vicariously accountable for what their algorithms do, the law could promote efficient incentives for people who develop or deploy algorithms and secure just outcomes for victims.
The challenge is to find a model of vicarious liability that is up to the task. This Essay provides a set of criteria that any model of vicarious liability for algorithmic harms should satisfy. The criteria cover a range of desiderata: from ensuring good outcomes, to maximizing realistic prospects for implementation, to advancing programming values such as explainability. Though relatively few in number, the criteria are demanding. Most available models of vicarious liability fail them. Nonetheless, the Essay ends on an optimistic note. The shortcomings of the models considered below hold important lessons for uncovering a more promising alternative
On Facial Recognition, Regulation, and Data Necropolitics
No full textThis paper argues for actual and legal regulation of artificial intelligence (AI) and facial recognition. These new technologies represent great opportunities to improve the welfare of societies. However, some of their uses can also enhance discrimination and, eventually, lead to violence. From a comparative approach (examining the European Union and Brazil), we address the current and future aspects of facial regulation, AI, and personal data. This paper shows that regulation is relevant to protect the rule of law, free markets, and individual freedoms. It also examines the looming risks unfolding from the unregulated uses of new technologies. Our concept of “Data Necropolitics” defines a predatory form of digital governance that exploits and discriminates against vulnerable populations
Clifford Awarded Ostrom Fellowship
Get PDFA 2L from Indianapolis has been awarded a prestigious graduate fellowship from The Ostrom Workshop at Indiana University.
Nick Clifford will begin a one-year Ostrom Fellowship in Fall 2023
The Contentious Issues of Governance by Algorithms
No full textThe development of computerized tools that lead to decision-making processes which apply locally defined parameters poses many questions about democracy. These questions stem from our very conception of the state and its role, going beyond the boundaries of typical administrative law. According to a popular notion that permeates the practices of most executive branches in liberal political regimes, democratic concerns are now competing with managerial concerns. In order to analyze this idea, we must study the implementation of algorithms in administrative decision-making, underscoring both the changes to the characterization of administrative decisions and the questions raised about an administrative judicial review of litigation. To summarize a French administrative law judge’s review so far, the judge began by assessing the legality of using algorithms in administrative procedures. Secondly, the judge reviewed the legality of making administrative decisions on the basis of an algorithm. Three issues now appear to be guiding the future of algorithm-based administrative decisions: (1) the security of legal transactions; (2) the compensation for harm or damage caused by the algorithms, and (3) the degree of in-depth review by the administrative judge
Four Maurer School of Law Students Selected as 2023 Stevens Fellows
Get PDFFour Indiana Law students have been selected as Stevens Fellows, the John Paul Stevens Foundation accounced today (June 20). Selection as a Stevens Fellow allows students to receive critical financial support while participating in unpaid summer legal internships serving the public interest.
Named after the late U.S. Supreme Court Justice, the John Paul Stevens Foundation is dedicated to promoting public interest and social justice values in the next generation of American lawyers
Avoiding Scandals Through Tax Rulings Transparency
Get PDFIn 2014, the International Consortium of Investigative Journalists broke the “LuxLeaks” scandal, revealing numerous tax rulings that the press termed “sweetheart deals” granted to multinational companies. Many countries offer tax rulings because they provide certainty to taxpayers and the government on the tax consequences of a planned transaction. Yet, secrecy that is followed by leaks and criticism is a recurring aspect of these rulings, both in the United States and Europe. LuxLeaks, which revealed secret rulings from the small European country of Luxembourg, was international headline news. It helped trigger widespread reforms. Tax authorities, including those of European countries and the United States, now automatically share information about cross-border advance rulings with other countries’ tax authorities. But Luxembourg’s tax rulings otherwise remain confidential. The United States treats a type of tax ruling, the Advance Pricing Agreement (APA), similarly: it exchanges information about APAs with other countries but does not otherwise disclose them. How transparent should tax rulings be? Secret rulings protect taxpayer confidentiality but also impose costs on various stakeholders. This Article (1) draws on the repeated scandals involving tax rulings to develop an original typology of these costs; (2) catalogues the levels of possible rulings disclosure, connecting each level with the costs it would address; and (3) examines potential arguments against rulings transparency. The Article concludes that, despite government resistance, best practices call for public disclosure of anonymized tax rulings—both letter rulings and APAs—heavily redacted, if necessary
Guilty of Probable Cause: Public Arrest Records and Dignity in the Information Age
Get PDFThe United States is exceptional among Western nations in its treatment of criminal records. Today, an estimated one-third of Americans1 bear the “modern equivalent of branding”: the publicly-accessible criminal record.2 Far from remaining locked in digital limbo, these records serve a variety of purposes, from legitimate law enforcement use to extortion against arrestees seeking to scrub their mugshots from a Google search.3 It would be natural to assume that such records result from an individual’s commission of a crime, for which the individual is duly convicted and then marked with the brand of the state for the transgression. But the scarlet letter of criminality enshrined in a record is often imposed in the absence of formal conviction, in the form of an arrest record. The widespread availability of these records leads to damaging collateral consequences for arrestees. This note will argue that a foundational interest in dignity, more prominent in Western European and international law than in the United States, can be a meaningful driver for criminal record reform in the United States. This paper will proceed in three parts. Part I will examine the treatment of criminal records in the United States, specifically those harms resulting from the widespread public availability of arrest records. Part II will examine Western European and international approaches to criminal records and how a robust interest in dignity has mitigated many of the issues faced in the American system. Part III will examine the opportunity for dignity to become a meaningful foundation for implementing criminal record reform in the United States
Design Patent Nonobviousness: The Road Ahead?
Get PDFIn this Essay, I comment on Mark Bartholomew’s Article Nonobvious Design and link it to an important contemporary debate over the nonobviousness test for design patents in the en banc rehearing in LKQ. I begin by challenging the assertion that nonobviousness doctrine is the most compelling vehicle for design patent policy reform, suggesting that Nonobvious Design may be borrowing this thinking from utility patent law, where nonobviousness plays a different role. I then turn to Nonobvious Design’s elaboration of the concept of the “aesthetic middle,” which provides the chief impetus for the suggested nonobviousness reforms. I question the normative implications of using the aesthetic middle as the driver of patentability for designs. Finally, I take up the reforms to design patent nonobviousness suggested in Nonobvious Design, and those at issue in the LKQ case, and offer my views on how the court should rule in that case
The Hidden Cost of University Patents
Get PDFUniversities are encouraged to undertake research through grants from government agencies, foundations, and other organizations. The Bayh-Dole Act reinforces this incentive structure by allowing universities to take ownership of the resultant patents. The rights of ownership include the ability to generate income by licensing patents and bringing patent infringement lawsuits. Undoubtedly, exercising these rights to financially benefit the university is economically rational. But might such actions also impose a cost on the public despite the fact that these very patents arose from public research subsidies?
This study examines the relationship between a university’s research expenditures and its likelihood to litigate patent infringement claims. It finds that research expenditures increase litigation frequency, suggesting that universities may use funds earmarked for research and innovation on patent litigation. We argue that patent rights provided by the Bayh-Dole Act may motivate this phenomenon—which encourages universities to seek rents, rather than pursue innovation. Our study adds to the extant literature about firm behavior, describing universities as vertical integrators as well as horizontal coordinators. It further suggests that this coordination inures to a university’s private benefit—but not necessarily the benefit of the public, for which universities are ostensibly organized