A Bubbling Dispute: Litigation Research at the Crossroads of Artificial Intelligence and the Law

By The Colleges of Law
Smiley faceThe Colleges of Law - Wed, Dec 27, 2023

Over the next several months of Legal Currents and Futures posts, The Colleges of Law faculty and students are sharing a series of thought pieces about artificial intelligence and the law. This article was penned by Nicole Clark.

For discussion purposes, let us begin with an exemplar case that will illustrate how legal analytics relies on AI, and specifically machine learning, to shape our litigation strategy. The facts of the case involve a dispute between local residents and Santa Barbara County and are as follows. The Hot Springs Canyon Trailhead sits alongside the foothills at East Mountain Drive and Riven Rock Road in Montecito. The trail ushers hikers past one mansion after another, guiding them into the Santa Ynez Mountains, where geothermal sulfur springs await a relentless stream of visitors. Over the past few years, these visitors, parked helter-skelter in lines stretching down each road, have led residents to install various barriers—boulders, signs, trees—on the street’s right-of-way. These encroachments, in turn, shoved parked vehicles into travel lanes, effectively transforming portions of a dual-lane road into a single lane while plunging the County of Santa Barbara into the courtroom.[1]

Backgrounding AI

Before moving on to legal analytics related to generative AI, a brief connection to recent advances in this transformative technology will be useful.  On November 30, 2022, OpenAI unleashed ChatGPT into the wild. The idea of a chatbot, a computer program designed to simulate human conversation, has existed since the 1950s.[2] This particular chatbot, however, has transformed how the general public interacts with artificial intelligence (AI), especially its generative capabilities. Generative AI can absorb and assimilate vast information and craft fresh, unique content. This power should make every attorney pause.

A significant portion of an attorney’s responsibilities involve written communication. We compose emails, draft memorandums, author briefs, and manage transactional records. “[Language is] really the currency that folks trade in,” said Daniel Katz, a law professor at Chicago-Kent College of Law. “Every road leads to a document. Either you have to read, consume, or produce a document.”[3] Still, the current fascination with new forms of generative AI overlooks the extent to which AI has been part and parcel of everyday legal practice for decades.

My company, Trellis has spent years compiling an archive of state trial court records, building an analytics platform that brings together the raw, messy data from courthouses across the country, restructuring and reclassifying the information with natural language processing and machine learning technologies to create unique, usable datasets. All of this work required AI. It required models that normalized the frequent misspellings of law firm names, identified the differences between two different judges with the same name, and categorized filings into specified practice areas. Trellis, and other legal analytics tools like it, rely on AI that sit under the radar. Researchers using software never see this AI, but it makes legal analytics possible.

A Spring of Fresh Insights

Returning to our case example will illustrate both the function of legal analytics and the role of AI in providing strategic litigation insight. On June 27, 2023, Hon. Thomas Anderle of the Santa Barbara Superior Court ruled in favor of the residents at the Hot Springs Canyon Trailhead in Montecito. According to his ruling, the County of Santa Barbara violated the letter and the spirit of a preliminary injunction when it placed notices in front of eleven homes, instructing residents to remove the encroachments they had installed within the public right-of-way on East Mountain Drive. The preliminary injunction, issued by the Hon. Donna Geck, was intended to block the County of Santa Barbara from moving forward with a parking project planned for the trailhead until it had conducted an environmental impact report in accordance with the California Environmental Quality Act.[4]

I found this ruling, along with its docket and all of its associated documents, using AI on a legal analytics platform. I typed hot springs canyon trailhead into the platform’s search bar, which immediately brought me to Case No. 22CV01299. The natural language models built into the search engine eliminated the need for more cumbersome Boolean search queries by working out the meaning of the whole query before producing a result in that context.

With access to the court docket, I can now see the names of all the attorneys associated with this case. When I click on one of these names, I am brought to a case list, where I can browse an archive of other cases touched by that same attorney. This is the building block for any type of law firm analytics. I can follow the Hot Springs Canyon Trailhead attorneys across time, watching as they move from one law firm to another. And, when I do follow these attorneys, I learn that one has belonged to a law firm with extensive experience before the Hon. Thomas Anderle; thirty-five percent of the firm’s Santa Barbara County cases happened within his courtroom, a number that suggests a deep familiarity with his judicial praxis.

By browsing through this litigation history, attorneys can glean new insights about other law firms and their place in the market, explains Stephen Embry, a former member of Frost Brown Todd.[5] Law firm intelligence makes possible new forms of quantitative and qualitative analytics, enabling an attorney to ask new kinds of questions about the legal landscape. In the case of Hot Springs Canyon Trailhead, there’s more I want to know: To what extent have these attorneys built their careers around natural resources litigation? Who are their largest, or most litigious, clients? Do they tend to represent individuals, government agencies, or Fortune 500 companies? All of these questions can now be answered in minutes.

The Expanding Headwaters

Folklore about everyday legal practice is filled with stories about attorneys asking their colleagues for anecdotes about a judge, a jurisdiction, a client, or an opposing counsel—anything that might help them litigate a case. An attorney needs information that extends beyond the boundaries of case law. To help meet this need, an increasing number of attorneys are turning to AI to perform this type of research, scaling their legal analytics capabilities with the deployment of API-backed case management systems.

But what is an API? APIs are everywhere. You use them every single day without even knowing it. An API, or an application programming interface, allows software developers to programmatically interact with software resources outside of their own codebase. What does this mean for an attorney? Legal analytics platforms can make portions of their database accessible to clients through APIs, and, as the years unfold, more and more law firms will capitalize on this offering. Why? With API access, law firms can automate the consumption of state trial court data to enhance their information workflows and reporting. 

Imagine the following scenario. An attorney is preparing a CEQA action against the County of Santa Barbara. She stumbles across the Hot Springs Canyon Trailhead case, noticing similarities in the legal issues at hand. She continues her search and identifies nine other cases broaching the same legal questions. This attorney might then ask her paralegal to visit the court’s website twice a day to see if there is anything new about these ten cases. This paralegal, who is fielding similar requests from multiple attorneys within the firm, is pressed for time. She could input each case into the court’s website twice daily, scrolling through an ever-growing docket to manually report on any updates. Alternatively, she could use an API to schedule automated searches through the ten case dockets, instructing the API to integrate any new information into the firm’s case management system. What would you like her to do?

Generating New Futures

On July 19, 2023, the California Court of Appeals published a ruling in favor of the County of Santa Barbara, finding that the Santa Barbara County Superior Court abused its discretion when it issued a preliminary injunction prohibiting the Road Commissioner from removing unpermitted encroachments on the public right-of-way along East Mountain Drive.[6] According to the appellate court, the installation of unpermitted encroachments is a misdemeanor, the management of which may not be linked to the California Environmental Quality Act.

As a case study, the Hot Springs Canyon Trailhead is a refreshing reminder that integrating AI into the litigation landscape does not mean that law firms will suddenly have to become software companies—that attorneys will suddenly have to become software developers. But it also serves as another reminder. Just as AI has bolstered legal research, it might soon do the same for legal writing and analysis, expanding the boundaries of what it might mean to practice the law. As generative versions of AI continue to improve over the next few years, we may see more and more attorneys harnessing new technologies to generate first drafts of complex legal instruments, to communicate with clients through natural language question-answering systems, or to produce more accurate predictions of legal outcomes—all of which are the very essence of legal practice. What will the law look like then?

Nicole Clark is the CEO and co-founder of Trellis, an AI-powered legal research and analytics platform that gives state court litigators a competitive advantage by making trial court rulings searchable, and providing insights into the patterns and tendencies of your opposing counsel, and your state court judges. Clark is a business litigation and labor and employment attorney.

[1] Yamamura, Jean. “Santa Barbara County Scores Big Win in Lawsuit Brought by Hot Springs Neighbors.” Santa Barbara Independent, 29 July 2023, <https://www.independent.com/2023/07/29/santa-barbara-county-scores-big-win-in-lawsuit-brought-by-hot-springs-neighbors/>.

[2] Bayerque, Nicolas. “A Short History of Chatbots and Artificial Intelligence.” Venture Beat, 15 Aug. 2016, <https://venturebeat.com/business/a-short-history-of-chatbots-and-artificial-intelligence/>.

[3] Ryan-Mosley, Tate. “AI Might Not Steal Your Job, but It Could Change It.” MIT Technology Review, 3 April 2023, <https://www.technologyreview.com/2023/04/03/1070750/ai-jobs-legal-field-gpt-4/>.

[4] Anderson v. Cnty. of Santa Barbara (Cal. Sup. Ct. 2023). <https://trellis.law/ruling/22cv01299/christopher-anderson-v-county-santa-barbara/20230503853802>

[5] Embry, Stephen. “Trellis New State Court Analytics Tools: Improved Litigation Decisions and Better Firm Management.” TechLaw Crossroads, 31 July 2023, <https://www.techlawcrossroads.com/2023/07 /trellis-new-state-court-analytics-tools-improved-litigation-decisions-and-better-firm-management/>.

[6] Anderson v. Cnty. of Santa Barbara (Cal. Ct. App. Jul. 19, 2023).