AI and the law
Henry A. Thompson · 2024
Read AI and the law by Henry A. Thompson (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2412-05090.txt.
Introduction
Henry Thompson, writing solo in 2024, argues that generative AI will reshape the law unevenly across its branches. He treats the technology as a labor-augmenting tool that cuts the cost of both writing fuller contracts and fighting cases in court. Those two effects pull in opposite directions: fuller contracts need courts less, while cheaper litigation draws more disputes in.
Where contracts are common, as in property and contract law, the net change in litigation is unclear because the effects offset. Where contracts are rare, as in tort law, litigation should rise. Following classic work from the 1970s, Thompson argues that this pressure will speed tort law's drift toward efficiency.
Catalog record
| CREATOR | Henry A. Thompson |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 50 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2412-05090
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
AI and the law Henry A. Thompson Abstract I argue that generative AI will have an uneven effect on the evolution of the law. To do so, I consider generative AI as a labor-augmenting technology that reduces the cost of both writing more complete contracts and litigating in court. The contracting effect reduces the demand for court services by making contracts more complete. The litigation effect, by contrast, increases the demand for court services by a) making contracts less complete and b) reducing litigants’ incentive to settle, all else equal. Where contracts are common, as in property and contract law, the change in the quantity of litigation is uncertain due to offsetting contracting and litigation effects. However, in areas where contracts are rare, as in tort law, the amount of litigation is likely to rise. Following Rubin, 1977 and Priest, 1977 , generative AI will accelerate the evolution of tort law toward efficiency. 1 Introduction Scholars have long recognized that technological progress shapes the evolution of the common law. For example, the proliferation of rail and heavy machinery during the Industrial Revolution spurred the shift from strict liability to negligence by increasing the number and severity of accidents ( Posner, 1972 ; Gifford, 2018 ) . 1 1 1 Fleck and Hanssen, 2024 also study the railway boom. They find that the growth of rail encouraged a shift away from a property rule called the benefit offset that helped railroads reduce payments for land acquired through eminent domain. In the 20th century, increases in the complexity of production drove the rise of strict products liability ( Landes and Posner, 1989 ) . Technological improvements at home and at work propelled the no-fault divorce revolution of the 1970s by increasing women’s returns on the labor market ( Leeson and Pierson, 2017 ) . 2 2 2 See also Friedman, 1984 . Such work shows that whenever new tech switched the least-cost avoider or highest-valued user of a property right, the law evolved with it. I consider how the law will change in response to an altogether different kind of innovation. Generative AI is a labor-augmenting technology that is especially likely to change the costs of litigation rather than the least-cost avoider. Indeed the most striking aspect of generative artificial intelligence (AI) is how much it boosts user productivity. Preliminary research has found that generative AI boosts materials discovery by 44% ( Toner-Rodgers, 2024 ) , programming speed by 56% ( Peng et al., 2023 ) , customer support agent speed by 14% ( Brynjolfsson et al., 2023 ) , writing task completion speed by 40% ( Noy and Zhang, 2023 ) , and consulting output and quality by 12% and 40% ( Dell’Acqua et al., 2023 ) . Choi et al., 2023 find that generative AI boosts legal task completion speed by 12-32%. In this article I explore the implications of such advances for the law. I argue that generative AI will alter the speed of the law’s evolution in two ways. First, generative AI will help attorneys write more complete contracts, weakening the demand for courts. Second, generative AI will reduce the cost of litigation, thereby boosting the demand for courts. When inefficient rules are litigated more often than efficient ones, these changes in the demand for courts will change how quickly the law evolves ( Priest, 1977 ; Rubin, 1977 ) . At first glance, the net effect of generative AI on the law’s rate of change in totum is ambiguous. The two effects offset one another. However, each effect will not be present in every area of law. In tort law, legal change will accelerate. This is because there are no contracts that would reduce the demand for courts. As a result, cheaper litigation will increase the rate at which inefficient allocations of tort liability are challenged and ultimately overturned ( Priest, 1977 ; Rubin, 1977 ) . In property and contract law, the contracting effect and the litigation effect work in opposite directions. Although generative AI incentivizes more litigation due to lower litigation costs, it also incentivizes less litigation thanks to more complete contracts. The net effect depends upon how elastic the marginal benefit of gap-filling is relative to the marginal cost of gap-filling. The demand for courts and legal change is likely to fall on net where the marginal benefit of contractual gap-filling is more elastic. Examples include commercial real estate and major infrastructure projects where even minor problems can have uniformly high costs. However, the demand for courts and legal change is likely to rise on net where the marginal costs of contractual gap-filling are more elastic as in commodity trading markets, mass consumer goods, and markets in which most participants are retired. My article contributes to three branches of scholarship. The first branch studies technology-driven institutional change. There, scholars have studied how technology sparks changes in the organization, structure, and scope of government ( Batchelder and Freudenberger, 1983 ; Allen, 1998 ; Allen, 2011 ; Allen and Barzel, 2011 ) or religious institutions ( Rubin, 2014 ) . 3 3 3 Other papers have studied how institutions slow or speed up the adoption of novel technologies ( Coşgel et al., 2012 ; Allen and Leeson, 2015 ) . The second is the now substantial body of scholarship on the evolutionary efficiency of law inculcated by Priest, 1977 and Rubin, 1977 . 4 4 4 See, for instance, Parisi, 2001 ; Parisi, 2002 ; Fon and Parisi, 2003 ; Fon and Parisi, 2006 ; Albrecht et al., 2022 . The paper also contributes to the growing body of work on AI in the law. The predictive ability of AI features prominently there ( Alarie et al., 2018 ; Gans, 2024 ) . 5 5 5 For work that examines the regulation of AI, see for instance Arbel et al., 2024 Of the work on AI in the law, the papers nearest to my own are Gans, 2024 and Casey and Niblett, 2020 . Gans, 2024 examines the demand for AI when it can produce accurate predictions about the outcome of a trial. Casey and Niblett, 2020 evaluate the effect of “robot judges” on litigation and settlement outcomes. Neither consider the implications of generative AI for contract completeness, attorney productivity, or the evolution of the law. 2 Context Attorneys produce and sell information about property rights. 6 6 6 For a slightly different perspective, see Gilson, 1984 who suggests that a business attorney’s main purpose is to reduce information asymmetries between the two parties. For example, before a dispute occurs and where a contract exists, attorneys produce information about the seller’s current assets and liabilities and a buyer’s rights and future obligations. Attorneys also produce and sell information after a dispute has occurred, whether a contract exists or not. In that case, they sell information about the law, regulations, precedents, sound legal arguments, and procedural rules that are beyond their clients’ ken. To produce information about property rights, an attorney spends a great deal of time either reading or writing. The three main tasks of an attorney are legal search, creating legal arguments, and legal writing ( Tu et al., 2023 ) . Legal search involves