[co-authors: Marc E. Williams, Anna J. Williams]*
The image of a robot takeover once existed only in media and literature. Now, fear of robot-like machines continues to creep its way into the professional realm with the ever-innovating field of artificial intelligence (“AI”). Although the true extent of the impact of AI on the legal profession (and the overall global labor market) remains unclear, the technology already possesses a significant ability, for better or for worse, to alter existing professional norms. While writing this article, I asked ChatGPT, “What will be the impact of artificial intelligence on the practice of law?” The machine answered, “Artificial intelligence (AI) is already having a significant impact on the practice of law, and its influence is expected to continue growing in the future.” In its full response, the machine listed all major points that I sought to raise in this article about AI’s impact: automation of legal work, the increasing value of efficiency, AI as a transformative technology, and accompanying ethical concerns raised by AI systems. In effect, my conversation with the machine about the impact of the machine underscored its increasing capability to place itself directly within the legal profession.
AI describes a massive body of technology geared toward training computer systems or machines to perform tasks that ordinarily require human intelligence. (Prashnu Verma and Rachel Lerman, A Curious Person’s Guide to Artificial Intelligence, Tech (May 7, 2023). Concerns over AI’s impact on the legal profession escalated with the advancement of large language models like ChatGPT. (Lee B. Ziffer, The Robots are Coming: AI Large Language Models and the Legal Profession, American Bar Association, Practice Points. Feb. 28, 2023). Large language models (“LLMs”) produce human-like responses based on the input of massive text-based datasets. Because LLMs only become smarter and more refined with every input, weary onlookers fear that the machine will inevitably replace human intelligence.
If you are a customer of Amazon, you are subject to their AI commerce platform every time you log onto their website. The Amazon system is designed to learn your buying preferences and present you with options consistent with your prior buying decisions. Likewise, Facebook uses an AI interface that presents paid ads in your timeline taken from the subjects of your search engine entries. If you search for information on desktop printers, voila! An ad for a printer will appear in your Facebook timeline. This is AI.
Fears of robot takeover are misplaced. Even though AI is rapidly developing, the technology has already proven to be incredibly useful, rather than ruinous, to various professions – including the legal field. For instance, Technology Assisted Review of eDiscovery is a product of AI that expedited a once-tedious task reserved for new associates or contract lawyers. While TAR (or any present AI system) is not foolproof, its profound impact on expediting and expanding the process of discovery cannot be overstated. (Myths and Facts about Technology Assisted Review, https://legal.thomsonreuters.com/en/insights/articles/myths-and-facts-about-technology-assisted-review).
TAR is not the only example of disruptive legal technology that changed the legal landscape. In fact, disruptive legal technologies that emerged in the 20th century alongside the development of the computer have consistently and profoundly impacted the legal profession in many forgotten ways. A History of How Technology Has Transformed the Legal Field, E-discovery 101 (Sep 9, 2021), For instance, for many decades, law firms and clients have used predictive analytics, or data points such as costs and outcomes of similar cases to determine how to evaluate the efficiency of case-resolution. In the 1970s, digital legal research eliminated the traditional, time-consuming practices of print-based legal research. In the 1990s, the legal profession started using online dispute resolution and e-Filing systems to keep up with the explosion of connectivity and information. At the turn of the century, the profession responded to the technology boom by automating case management, time-tracking, and billing. As the 2000s progressed, so did digital productivity tools, legal apps, video conferencing, and blockchain contracts. Now, as a positive product of the COVID-19 pandemic, case handling, depositions, mediations, hearings, and even trials are handled remotely in ways that were once considered impossible. Many of these now ordinary digital products were at one time viewed with harsh skepticism.
Accordingly, simply because AI exists as an abstract concept to most does not render it dangerous or useless to lawyers. Rather, with basic human understanding, AI can become a competitive tool that automates mundane tasks and advances legal practice areas by forcing the industry to meet new market demands for specialization. While there is little doubt that AI disrupts the existing norms of the legal profession by catalyzing the demise of the billable hour and generalized practice, lawyers must adapt alongside the technology to improve their expertise, output, and client relationships.
A technological transformation in the legal industry is afoot. To survive it, the legal profession must examine its practices and requisite professional skills. This article takes a multifaceted approach toward analyzing the impact of AI on the civil defense lawyer. First, this article will examine AI’s role as a modern disruptor to a traditional profession predicated on exclusivity. Then, this article will analyze two key products of AI’s disruption: the demise of the billable hour and changing professional demands. Finally, this article will conclude with a series of solutions to quiet the disruption and adapt with automation.
AI as the Modern Disruptor
Generative AI is the new technological disruptor in the legal profession. This broad, rapidly developing category of technology can produce documents, drafts, and data only seconds after a basic human input. The hallmark feature of generative AI is its ability to create text, sound, or images based on minimal human prompts and constantly improving, trained data sets. For instance, the technology can generate English essays, professional headshots, and even covers of songs by an artist of choice. (Verma and Lerman). ChatGPT and similar software fall within the category of generative AI.
What makes generative AI so disruptive is its potential to expedite human processes and, for some, its potential to replace the need for human intelligence. Before GPT, someone writing an essay might conduct cursory research on a topic, outline the topic, and create several drafts before completion. With generative AI, this same person can invest a miniscule fraction of that time into the project by simply generating a series of prompts. Generative AI enhances efficiency tenfold.
However, efficiency comes at a cost to professionals. Professionalism relies on exclusivity. In other words, a professional must have some set of trained, special skills not available to a lay person to make them useful in the labor market. For lawyers, a significant portion of their exclusivity rests in their research, analytical, and writing skills. Of course, access to exclusive knowledge and skills comes with a price tag. And, naturally, where there is no shortage of demand, the supplier – lawyers – can control the price. Economic leverage controls the profitability of the legal system, but generative AI poses a significant threat to it by bringing efficiency to the forefront. (Rebecca J. Kunkel, Artificial Intelligence, Automation, and Proletarianization of the Legal Profession, 56 CREIGHTON L. REV. 69 (2022)).
With efficiency at the forefront, recipients of costly professional services are thinking differently. Not only can generative AI models like ChatGPT expedite professional services, but these models can also improve accessibility to the information, documents, and processes professionals follow. To be clear, this does not mean that increasingly capable machines create decreasingly necessary professional intervention. Rather, these machines reshape and redefine traditional services, such as legal services, into a more-for-less model. (Richard Susskind, Tomorrow’s Lawyers: An Introduction to Your Future (3d ed. 2023)). Lawyers can do more in less time than ever before, which in turn means that clients can access more and, perhaps, pay less for that work product. This has the potential to revolutionize the pricing of legal services, and perhaps, lead to (finally!) the demise of the billable hour.
The lure of efficiency is tempting, even to a professional. However, the unbridled use of generative AI in the legal profession poses significant ethical concerns that, without proper understanding, create dire consequences for clients and in courtrooms. For instance, Steven Schwartz, a lawyer in New York, fell victim to ChatGPT’s occasional hallucinations, or output of false information, in federal court. In the case of Roberto Mata v. Avianca, Inc., Schwartz cited six cases generated by ChatGPT during his legal research. A legitimate citation followed each case; however, it was soon discovered by opposing counsel and the court that these cases, and citations, did not exist. Schwartz, in an affidavit responding to the Court’s Order to Show Cause, admitted to using the research tool and promised to never use it again without verifying the authenticity of each source. Stephen Schwartz, his co-counsel Peter LoDuca, and their law firm Levidow, Levidow & Oberman were ordered to pay a $5,000 fine for “acts of conscious avoidance and false and misleading statements to the court.” (Sara Merken, New York Lawyers Sanctioned for Using Fake ChatGPT Cases in Legal Brief, (June 26, 2023, 4:28 AM), https://www.reuters.com/legal/new-york-lawyers-sanctioned-using-fake-chatgpt-cases-legal-brief-2023-06-22/).
Ultimately, what happened to Schwartz is a common fault of the current form of generative AI. The machine is trained to produce responses based on a prompt. Schwartz most likely asked ChatGPT to give him cases applicable to a favorable proposition in his case. Accordingly, the machine analyzed its databank of case law, likely found on the internet, and produced a response with only the favorable, most applicable portions of each case. The efforts of the lawyer to find an easy solution to a common lawyer task (legal research) ended up costing him not only the case and a monetary sanction, but his credibility before any judge where he practices.
Generative AI poses several other risks to the legal profession and to most any subject area it touches. First, the system inherits the bias of the humans that train it with their inputs. As of now, generative AI lacks the capability to discern harmful human thought patterns. This is partly because humans themselves struggle to understand their implicit biases. Even so, training generative AI on inherently flawed information can exacerbate implicit biases. For instance, in 2016, Microsoft launched its Twitter AI-based chatbot named Tay. Microsoft intended for Tay to be a playful internet chatbot which would respond to user’s prompts. But less than 24 hours after going live on the internet, Microsoft removed Tay, as it had devolved into what many see too commonly on social media: a hate-filled, antisemitic, racist, and hate-spewing robot. (Microsoft Chatbot is Taught to Swear on Twitter, (Mar 24, 2016), https://www.bbc.com/news/technology-35890188).
Aside from being an inherently flawed model, generative AI may also pose a privacy risk to legal professionals and their clients. For instance, one should never put client confidential information into a Chat GPT inquiry, as the system will utilize all data fed into it in answering other inquiries. Unless the user has the forethought, understanding, and option to change data and privacy settings before using the system, everything put into a generative AI system stays in the system to train a better response for the next input. Lawyers could potentially divulge confidential information in their quest for the perfect paragraph or absolute answer. Lawyers could accidentally upload confidential files into a public databank when seeking a simple proofread or argument enhancement. The privacy concerns are great.
The final remaining ethical challenge in the age of generative AI is in assigning value to legal work product. If the system is truly capable of altering the practice of law by automating nearly half of it, then lawyers must reconsider what is a reasonable fee in light of what the system can produce.(Amy B. Cyphert, A Human being Wrote This Law Review Article: GPT-3 and the Practice of Law, 55 UC Davis L. REV. 401 (2021)). This ethical reconsideration is perhaps the start of a larger conversation on billing, which we address later in this article. Needless to say, generative AI will force a change within many areas of the legal profession.
Despite developments in generative AI that dramatically enhance efficiency, data about lawyers and their legal research tendencies tells a different story about its present utility. According to the ABA’s 2022 Profile of the Legal Profession, lawyers are spending more time conducting legal research than in previous years. Nearly half of lawyer’s report that their research begins on free online search engines like Google. However, only 10% currently use the artificial intelligence tools in their firms. Interestingly, the number increases to 19% at larger law firms with more than 100 employees. (American Bar Association, ABA Profile of the Legal Profession, (2022)). In April of 2023, Thomson Reuters issued a report on ChatGPT and its use in law firms. The report demonstrates the legal profession’s predicament: while 82% of lawyers agreed that generative AI can apply to legal work, only 51% agreed that it should be applied to legal work. (Thomson Reuters, New Report of Chat GPT & Generative AI in Law Firms Shows Opportunities About, Even as Concerns Persist, https://www.thomsonreuters.com/en-us/posts/technology/chatgpt-generative-ai-law-firms-2023/).
Still, despite this professional divide, companies continue to innovate what may become the new normal in the legal profession. For instance, in May 2023, LexisNexis announced its new generative AI platform Lexis+AI, a generative AI platform equipped to produce research, summaries, and documents. (Rhys Dipshan, LexisNexis Announces Generative AI Platform Lexis+AI, Automating Search, Drafting and Summary Tasks, (2023)). Shortly thereafter, Thomson Reuters announced its partnership with Microsoft to bring generative AI to its legal products, like Westlaw Precision, in 2023. (Matt Reynolds, Thomas Reuters Partners with Microsoft for Generative AI Push, (May 23, 2023)). In fact, other law firms already use generative AI in their practice. For instance, international law firm Allen & Overy launched Harvey in February 2023. Harvey is a legal-specific generative AI platform that assists its lawyers in conducting research with natural language instructions. (Caroline Hill, Allen & Overy Breaks the Internet (and New Ground) with Co-Pilot Harvey, (Feb. 16, 2023), https://legaltechnology.com/2023/02/16/allen-overy-breaks-the-internet-and-new-ground-with-co-pilot-harvey/). In the United States, Troutman Pepper launched its Generative AI Task Force in May 2023 to explore the use of generative AI within the firm and practice. (Troutman Pepper Launches Generative AI Task Force, (May 10, 2023). Practice specific generative AI avoids some possible errors because it is trained only on firm-data or accurate case law. However, even if a generative AI system reflects an intranet of information, it is unlikely, in its current model, to escape the occasional hallucination or output of misinformation.
The speed at which lawyers embrace generative AI in their practice may certainly change as the technology improves its accuracy and privacy functions and as the technology is embraced by subscription-based legal research platforms. But delay in the implementation of generative AI in the legal profession will not slow down its overall impact on lawyers. We must exercise our comprehension and critical thinking skills to answer the questions raised by generative AI and adapt to the professional opportunities it creates. If anything, generative AI forces the legal profession to confront some of its longest-standing systems and skillsets – the billable hour and the exclusivity of lawyering. In other words, in the age of AI and its accompanying emphasis on efficiency, lawyers must redefine how and for what services exactly they are billing. While the practice may be slow to adopt the technology, there is no doubt that it will force a rapid change in professional norms.
The Demise of the Billable Hour
Generative AI forces the legal profession to confront its standard billable hour model. In the background of the billable hour, the capability of generative AI to automate the legal profession creates a significant economic risk. Because the legal profession primarily relies on the billable hour model, which itself relies on the heavy-lifting of legal work being completed by less experienced associates, any time-saving technology inherently poses some risk to diminishing the total revenue generated in a law firm. But even in the age of AI, this proposition is not new. For decades, disruptive technologies like TAR, machine prediction, and research add-ons unsettled the billable hour model. However, none of these technologies possessed the production capabilities of generative AI.
Ironically, the billable hour originated from a desire to maintain efficiency and transparency between lawyer and client in an increasingly complex legal system. Early legal services were traditionally offered on a fixed-fee basis, billed at the conclusion of a matter. From the fixed-fee model, other payment arrangements like contingencies and retainers emerged. However, between the 1930s and 1940s, state bar associations struggled with stagnant attorney incomes, increasingly complex discovery requirements, and expanding federal regulations. Clients began equating the value of legal services with the length of time it took to resolve their legal matter. Major law firms sprouted across the country. All of these elements created the perfect window for the billable hour and its detailed record-keeping to rise in prominence. The increase in legal work, the length of time it took to complete that work, and the market competition raised incomes and revenues nationwide under the billable hour model. (Stuart L. Pardau, Bill, Baby, Bill: How the Billable Hour Emerged as the Primary Method of Attorney Fee Generation and Why Early Reports of its Demise May Be Greatly Exaggerated, 50 Idaho L. Rev. 1 (2014)).
Leverage is an integral characteristic of the profitability of the billable hour and the modern law firm. The system relies on fewer, more experienced lawyers who can filter work to more, less experienced lawyers. The billable hour model allows firms to generate revenue based on hourly rates multiplied by the time and effort spent on a case. Both the hourly rates and time spent often reflect the experience level of the lawyers at work. Both the costs saved, and revenue generated reflect a pyramid structure that filters more time-consuming tasks like research and drafting to lesser-experienced associates. Each of the elements combine to increase the income firm-wide. Each of the elements also necessitate detailed timekeeping and productivity tracking. Since the complexity of legal work increased in the mid-20th century, the billable hour has remained the dominate economic model for law firms and a preferred model for clients eager to track the resolution of their legal problems.
Despite its dominance, the model is a source of great criticism both within the profession, where it is viewed as a foe to efficiency, innovation and growth, and from consumers, who view it as incentivizing inefficiency and promoting a lack of access and cost transparency. Furthermore, a historic economywide drop in productivity, yet increase in legal work as a result of the pandemic has pushed many law firms to reconsider their pricing structures. (Andrew Maloney, The American Lawyer, As Productivity Drops, Will More Law Firms Move Away From Billable Hour?, (Jan. 18, 2023)). In fact, the request to use and the ultimate use of alternative fee arrangements, rather than the billable hour, rose during the pandemic. The 2023 Citi Hildebrandt Client Advisory Survey found that in 2022, large firms expected AFAs to account for 20.5% of their total revenue, a figure up 4.5% from 2016 (2023 Citi HildebrandtClient Advisory). Additionally, 39% of large firms expected that over half of their revenue in 2022 would come from AFAs and other pre-negotiated discounts. Alternative billing systems are providing modern firms more flexibility at work. Notably, these alternative billing systems focus on the quality of the output rather than the time input. Even so, Citi also concluded that the billable hour is not at its total end.
That is – until now. Generative AI changes the narrative and the controls. For consumers of legal services, disruptive technologies like generative AI are empowering. Legal consumers have greater access to the justice system than ever before. Albeit not at the moment, a future, improved form of generative AI can easily be used to formulate legal answers and documents that many consumers could not otherwise readily afford. Generative AI can also expedite the time-consuming aspects of the increasingly complex legal process. For the legal profession, this future requires careful consideration of changing client expectations.
Richard Susskind describes this problematic paradigm as the “More-for-Less Challenge.” (Suskind, Tomorrow’s Lawyers). In essence, there is no foreseeable decrease in the demand for legal services, especially in the increasingly digital age. However, technology shifts the expectations of those seeking legal services. If technologies like generative AI can make access to quality legal services much cheaper and much faster than ever before, very few clients, corporate or everyday citizens, will pass up that opportunity.
As always, skeptics caution that merely because something is cheaper or faster does not mean that the technology makes the service inherently better, but in a world of generative AI, this is simply untrue. The output produced by generative AI replaces the human knowledge behind the input with increasing precision. Accordingly, the human knowledge inputted into the machine becomes increasingly devalued as the machine becomes smarter. Thus, in time, the cheaper and faster option presented to clients may, in fact, be much more reliable than ever before.
This is particularly true where the generative AI model is not simply ChatGPT, but a system created by legal research experts with information exclusive to their subscription services. For instance, Lexis+AI is described as, “[A] generative AI platform designed to transform legal work. Lexis+ AI is built and trained on the largest repository of accurate and exclusive legal content, leveraging an extensive collection of documents and records to provide customers with trusted, comprehensive legal results with unmatched speed and precision and backed by verifiable, citable authority.” (Lexis Nexis Launches Lexis AI). Generative AI products produced by legal research companies can be trained on both factually and legally correct data sets that, for the time being, remain unavailable to free services like ChatGPT. While these machines may not be totally free of hallucinations and other research woes, there is certainly a higher degree of precision that lawyers can strategically adopt to streamline their output more efficiently than ever before. Not only can lawyers utilize the machines to conduct research and enhance comprehension, but lawyers can also have the machines automate the traditional drafting process while cutting back on many hours spent drafting and editing. Accordingly, client expectations will shift away from the traditional notion of legal services. Managing those expectations will require, at the very least, an explanation of why this technology is not being used.
Under the traditional billable hour model, the automation and efficiency potential produced by generative AI cuts against once reliable revenue streams and stifles the demand for much technical and associate-level work. For example, if you were researching the admissibility of scientific or technical evidence in a certain jurisdiction, that research could take an associate several hours to research the issue, read the cases and prepare a memo. Lexis+AI or Westlaw Precision, however, could provide case citations and write a memo summarizing the answer in a matter of seconds after entry of a simple, one-sentence inquiry. The lawyer would, of course, have to verify the conclusion. But comparing the time and cost of traditional research with the time to generate an AI conclusion is marked.
While automation does not replace legal work, it dramatically changes the value of it in the economy. For instance, in March 2023, Goldman Sachs issued a research report predicting the effects of AI on economic growth. On average, the report predicted that all industries would experience approximately 25% automation because of AI. But, when broken down by industry, the report predicts that 44% of legal work could be subject to automation because of AI. (Briggs,, and Kodnani, The Potentially Large Effects of Artificial Intelligence on Economic Growth, (Mar. 26, 2023), Much of that work is in commercial transactions, not tort litigation, but the ability to transform litigation through resolution platforms running AI data sets that analyze similar factual and legal situations make it likely that litigation as we know it will be forever altered by use of AI tools.
This statistic places the legal profession in the second-highest automation-risk category. However, automation does not necessarily implicate a full-scale replacement of lawyers. Rather, it requires lawyers to develop specialized skillsets and offerings to maintain the marketability of the costs of their services. It also requires that law firms explore creative, cost-effective ways to engage clients with either the implementation of new billing structures, like flat-fees and subscription arrangements, or the production of higher value legal services. Valuation, of course, belongs to a consumer most likely eyeing efficiency more than ever before. As such, high value legal services must either embrace AI by offering a technologically enhanced work product or provide for something that a machine cannot yet replace. In an upcoming age where legal information is readily and, for the most part, reliably available, it is the assigned value of legal work, particularly as characterized by the traditional billable hour model, rather than the work itself, that is truly disrupted.
Meeting Changing Professional Demands
Combatting the economic implications of generative AI’s automation requires that the legal profession differentiate its services through specialization and creative business models. The burden of differentiation will fall largely on young associates who typically find most of their work in traditional time-consuming tasks like document review and legal research. These tasks, if not already, will be dramatically expedited as the legal profession implements different modes of generative AI.
The concept of automation, albeit fear-inducing in many corporate conversations, is not wholly synonymous with replacement, decreased job prospects, or slowed job growth. Rather, automation in the generative AI sphere is an opportunity for specialization. Young associates looking to differentiate themselves should focus on a specialization and how it can relate to generative AI.
The legal profession has already experienced varying degrees of automation, many of which were most closely felt by younger associates. A prime example of this is the entire field of eDiscovery, which exploded in the middle of the 2000s. Before eDiscovery, young associates spent much of their early career combing for relevant documents in rooms full of banker’s boxes. But as technology boomed with the progression of the 21st century, the entire field of law confronted a massive explosion of data and, naturally, the complications associated with it. The field of eDiscovery emerged and, along with it, so did many new job specializations like eDiscovery specialists, data analysts, and program managers. Not only did the role of document review change, but so did the necessary credentials of those expected to perform it. The expansion of the field of eDiscovery also resulted in rapid market growth and profits. For instance, in 2022, the global eDiscovery market was valued at $11.2 billion. By 2027, it is predicted that the market will reach a value of $17.1 billion. (eDiscovery Market Statistics, (June 2022)). Rest assured, automation is not the end of the legal profession.
Accordingly, differentiation does not require completely redefining the role of an associate. Rather, differentiation requires that associates, and all lawyers, utilize disruptive technologies, like generative AI, as a tool to enhance their technological competence and sharpen their traditional lawyering skills. Rather than a method of replacement, generative AI enhances the capability of legal professionals who take the time to understand it. As a result, new, necessary credentials, job positions, and revenues will enter the field of law. Maintaining an advantageous position in the future labor market requires using and understanding generative AI.
Lawyers, particularly young associates, must become technologically competent to appropriately and successfully differentiate themselves in the changing workforce. It would be remiss to not acknowledge the dramatic impact that generative AI will have on traditional legal work. Generative AI is a streamlining tool that will replace a significant portion of time spent on client interactions, research, drafting, and review. However, if used correctly, the streamlining, or automation, effect of generative AI can increase workload capacity, accuracy, and overall legal knowledge. If anything, what generative AI replaces creates an opportunity for savvy associates to increase the quality of their output and their overall productivity earlier in their careers. Especially in the administrative sense, AI can replace nonbillable, tedious tasks while freeing up time to spend on billable work. The ability to devote extra time to billable work, along with a new set of tools brought about by generative AI, will most likely increase the quality of legal work beyond the typical measure of experience.
Furthermore, many young associates are products of a highly connected, technologically driven generation. Later Millennial’s and Gen-Z’s lifelong understanding of technology, and the speed at which they can adapt to its new versions, will be a great asset to their careers and to the practice of law, so long as ethical and privacy risks are not overlooked. Young associates should devote time and energy into understanding the strengths and weaknesses of each generative AI model before implementing it into their practice. By understanding both the strengths and weaknesses of generative AI, young associates can use it to enhance their work product and to maintain relevant among shifting workforce demands. A fruitful future in the legal profession depends on technologically competent associates who can understand, utilize, and innovate alongside generative AI.
As for the legal profession as a whole, lawyers of all experience levels must understand generative AI as neither a static nor linear concept. Generative AI is a form of automation that can adapt and integrate within the legal profession so long as members understand its purpose. Even in the digital age, it is remarkably unlikely that generative AI can replace the counseling and strategic decision-making skills of an attorney.
Even considering the impact of generative AI on the future of law, traditional advocacy skills are not fleeting. In fact, they may be more important than ever before. While generative AI can automate many areas of the legal profession, it cannot yet replace the necessary human component of meaningful advocacy and counsel. However, that is not to say that some creators have not tried. For instance, DoNotPay is an AI chatbot that provides legal services to consumers via mobile app. DoNotPay attempted to send an AI-powered chatbot into a courtroom to contest a consumer’s traffic ticket in February 2023. DoNotPay founder Joshua Browder abandoned the plan after being threatened with six months of jail time by “state bar prosecutors.” (Megan Cerullo, AI-Powered “Robot” Lawyer Won’t Argue in Court after Jail Threats, CBS News (Jan 26, 2023),
While the legal profession did not warmly embrace the idea of an AI-powered robot-lawyer in the courtroom, this situation exemplifies how the profession is navigating the boundaries of technology and lawyering. On the one hand, AI is incredibly useful to the research, writing, and management side of practice. On the other hand, AI in the courtroom is viewed as premature and dangerous. These are rational. AI is not yet able to train out the inherent human bias of those feeding it information. And while it may be able to replace some levels of thinking, it cannot entirely displace human nature, empathy, instinct, or rationality. The role of a lawyer as a counselor is invaluable. For these reasons, it is important that traditional advocacy skills are not forgotten as lawyers adapt to the age of generative AI.
Conclusion
Generative AI is the latest, but certainly not the last, disruptive technology to enter the labor market. While generative AI is still in its early stages, its future capabilities will force a change in the legal profession. These changes require the legal profession to make business adaptations to better serve an increasingly digital market. First, the legal profession must reexamine the utility of the billable hour in the efficiency-focused age of generative AI. Second, the legal profession must transform their skillsets to keep pace with changing professional demands. The value of a counselor is irreplaceable; however, a counsel in the age of generative AI must be adaptable to render useful advice. Ultimately, generative AI is as much as a disruptor as a transformer of the legal profession. The legal profession must set aside its skepticism and strategically embrace a future with generative AI to maintain their overall value.
* Marc E. Williams is the managing partner of the West Virginia office of Nelson Mullins. He is a fellow of the American College of Trial Lawyers and has a national trial and appellate practice involving commercial litigation, products liability, class actions, and mass torts. He is a past President of DRI, Lawyers for Civil Justice, and the National Foundation for Judicial Excellence.
Anna J. Williams is a third-year law student at the West Virginia University College of Law and a candidate for graduation in 2024. She is a graduate of Marshall University and spent the summer of 2023 as a Summer Associate in the West Virginia office of Nelson Mullins, which she will join as an associate upon graduation.
