New AI for Litigators

Editor’s Note: In this succinct exploration, Rob Feigenbaum and Patrick Oot wade into the expanding intersection of artificial intelligence (AI) and legal practice, particularly focusing on litigators’ application of AI tools to enhance their work’s efficiency and effectiveness. As AI’s footprint within the legal arena grows, its potential to transform traditional litigation workflows and methodologies becomes increasingly evident. This article surveys the capabilities and implications of AI tools such as Casetext CoCounsel and Prevail, highlighting how these innovations can streamline legal processes, from case research to court reporting and testimony management. Amidst technological advancements, the authors also address the paramount importance of cybersecurity, privacy, and the ethical considerations that accompany the integration of AI in legal proceedings. Through this examination of current AI tools and their practical applications, Feigenbaum and Oot offer a forward-looking perspective on how AI is set to redefine the landscape of legal practice, highlighting the balance between technological leverage and the indispensable human judgment of legal professionals.

Industry Expert Article*

New AI for Litigators

Rob Feigenbaum and Patrick Oot[1]

Over the past year, practitioners have predicted widespread adoption of artificial intelligence (“AI”) in the practice of law, and that AI will fundamentally change how litigators work.  See, e.g., Andrew Judkins, “Use of AI in litigation: A quick look at today and the future,” Norton Rose Fulbright (October 5, 2023) (“It is inevitable that the role of AI in litigation will continue to expand.  As with AI generally, the breadth and pace of change will be significant and reshape how litigation is conducted.”).

Litigators seeking to utilize AI to improve their efficiency and effectiveness have found a scarcity of real tools for implementation at the present time.  Does the emperor have no clothes? Casetext CoCounsel has emerged as the most prominent AI tool for litigators currently available.  Casetext permits a user to input basic information at a prompt, resulting in a report of legal research answers, citations and memos; summaries of key documents and testimony; chronologies and timelines; drafts of correspondence; extractions and analysis of contractual language; and other documents.  See www.casetext.com.  Of course, AI tools require a lawyer to review and edit the query and results, but the true benefit of current AI deployments is to support a process that speeds up the development of a basic framework for legal documents.  See John Rosenthal and Ashish Prasad, “Weighing AI’s and Risks in Litigation and eDiscovery,” Complex Discovery (November 8, 2023), https://complexdiscovery.com/weighing-ais-benefits-and-risks-in-litigation-and-ediscovery/.  These new AI tools free up litigators to focus on higher value tasks, and they can take on more work from clients due to the time savings resulting from using AI.

Another prominent AI tool for litigators now available is Prevail.  See www.prevail.ai.  Prevail is a court reporting and testimony management platform that harnesses AI to enable the following advances for litigators.

(1) Before a deposition or hearing, a litigator can access, search, summarize and identify key testimony from depositions and hearings by making plain English queries in the platforms rather than having to craft Boolean word searches across multiple repositories of testimony.

(2) During a deposition or hearing, in real time, the AI can identify whether questions in the litigator’s outline have been asked and then provide a link to where the testimony is in the deposition or hearing. This makes it much easier for the litigator to get through his or her entire witness outline for the examination.  Also, the AI tool links documents referenced during testimony in real time within the transcript.  One simply clicks the document to open it.

(3) During a deposition or hearing, in real time, potential inconsistencies between the testimony of the witness and that of other witnesses in the case, or the testimony of that witness or other witnesses in prior cases (if the testimony is part of the platform), are flagged. In addition, the platform will identify if the testimony relates to other documents in the case, and whether the documents potentially contradict the testimony. The litigator has the AI’s assistance in remembering and accessing the relevant prior testimony for the examination.

(4) During or after a deposition or hearing, a litigator can use the AI to summarize the testimony. This frees up the litigator, and/or junior litigators, from having to summarize the testimony, and yields cost savings for the client.

These features are all referenced on the Prevail website.

Very few litigators have photographic memories of their examination outlines, the past relevant testimony in the matter, and past relevant testimony from other matters.  Superhuman litigators are few and far between.  For most of us, the AI will be a welcome assistant in preparing for and taking witness examinations in depositions and hearings, and the AI will enable more efficient and effective witness examinations.

Also, Prevail provides a secure platform free from cybersecurity issues because, unlike traditional court reporting, it is SOC-2-Type 2 and ISO 27001 compliant, and it assures that the parties’ data and testimony (1) are always stored in, a secure cloud-based environment, (2) can be encrypted for use in transmission and storage, and (3) can be subject to regular third-party security audits, including penetration testing.  See Rob Feigenbaum, Geoffrey Vance and Patrick Zeller, Solving the Problem of Information Security in Court Reporting, Complex Discovery (January 16, 2024), https://complexdiscovery.com/solving-the-problem-of-information-security-in-court-reporting/.

Nor does Prevail raise significant concerns about invading privacy interests or compromise of legal privileges.  See ABA Model Rule 1.6© (lawyer must take reasonable efforts to prevent the “unauthorized or inadvertent disclosure” of information relating to representation of a client).  Like Casetext’s CoCounsel tool, Prevail uses dedicated services to access GPT-4, so that client data is not use to “train” the model and does not become part of publicly accessible knowledge.  See www.prevail.ai; www.casetext.com.

Finally, Prevail does not raise significant concerns about dehumanizing the law, hallucination, due process, reliability or potential bias.  See Rob Feigenbaum and Giel Stein, AI Tools for Judges, Complex Discovery (February 4, 2024), https://complexdiscovery.com/ai-tools-for-judges/ (citing See Chief Justice John G. Roberts, Jr., 2023 Year-End Report on the Federal Judiciary (December 31, 2023), pp. 5-7).  Like the technology platforms for electronic discovery, which have for many years identified for litigators the most important documents produced in discovery to assist litigators in preparing for depositions, hearings and trials, see www.relativity.com, Prevail simply gives litigators the ability to identify, quickly and efficiently, the most important testimony for use in the deposition or hearing.  Litigators must exercise their own judgment and decide what testimony is important, and should be referenced in a pleading, deposition or hearing.

Counsel is free to use technological tools, of course, but counsel must verify the results by checking their work.  A litigator thereby satisfies his or her duty of due diligence and competent representation of the client under the Federal Rules of Civil Procedure and ABA Model Rules of Professional Conduct.   See FRCP 11(b) (“[b]y presenting to the court a pleading, written motion, or other paper . . . an attorney or unrepresented party certifies that to the best of the person’s knowledge, information, and belief, formed after an inquiry reasonable under the circumstances: . . . (2) the claims, defenses, and other legal contentions are warranted by existing law or by a nonfrivolous argument for extending, modifying, or reversing existing law or for establishing new law . . . [and] (3) the factual contentions have evidentiary support . . .” (emphasis supplied)); FRCP 26(g) (“by signing [a request, response or objection], an attorney or party certifies that to the best of the person’s knowledge, information, and belief formed after a reasonable inquiry:  it is complete and correct as of the time it is made” (emphasis supplied)); ABA Model Rule 2.1 (““[i]n representing a client, a lawyer shall exercise independent professional judgment and render candid advice” (emphasis supplied)); see also ABA Model Rules 1.1, 1.6(a) and (c); and 3.3.  These rules require, among other things, reading the cases and transcripts before argument, or making filings with the court.  Trust, but verify, is the order of the day.

The new AI tools for litigators provide exciting opportunities for litigators to harness AI to improve their efficiency and effectiveness, and they enable litigators to provide better and more work for their clients.  We look forward to more AI tools emerging to help litigators in the years to come.

End Note

[1] Rob Feigenbaum is the Co-Founder and CEO of Prevail Legal.  Patrick Oot is Partner at Shook, Hardy and Bacon LLP.  The authors would like to thank Ashish Prasad, Vice President and General Counsel of HaystackID, for his valuable assistance in preparing this article.

* Shared with permission of authors.

Assisted by GAI and LLM Technologies

Additional Reading

• AI Tools for Judges
• Solving the Problem of Information Security in Court Reporting

Source: ComplexDiscovery OÜ