Does AI pose a threat to lawyers of tomorrow? Interview with authors of AI and Law: Challenges Demystified

ArtificiaI Intelligence and Law: Challenges Demystified by Rodney D Ryder and Nikhil Naren is a comprehensive textbook that explores the intersection of artificial intelligence (AI) and legal frameworks. The book addresses the complex challenges and opportunities in the rapidly evolving realm of AI..The book takes a deep dive into how AI is reshaping legal practices, from automated legal research to AI-driven contract analysis, while also raising important ethical, regulatory and privacy concerns. It also touches upon the role of AI in intellectual property, data protection, antitrust disputes and cybersecurity, and offers insights into how legal professionals can navigate this new landscape..Rodney D Ryder is the Founding Partner of Scriboard Advocates and Legal Consultants, a full-service commercial law firm with specialisation in technology, new media and IP laws. Ryder has over 25 years of experience in technology, data, privacy, IP and new media laws..Nikhil Naren is an Assistant Professor at Jindal Global Law School (JGLS) and Of Counsel at Scriboard. His research focuses on the intersection of law, technology and society..Edited excerpts of the interview follow..Jelsyna Chacko (JC): Are there any particular areas of law that are more vulnerable to disruption by AI and does this pose a threat to lawyers of tomorrow?.Nikhil Naren (NN): Chapters 2, 3, 4 and 5 of the book respectively talk about regulation, intellectual property rights and intersection of AI, the privacy issues and antitrust issues. There are of course certain areas of law that would end up being more vulnerable by the introduction of AI.For example, when Generative AI came into being, a year-and-a-half down the line, the floodgates of litigation have now opened because questions have moved beyond who gets to have the rights on intellectual property that is created by AI to whether an infringement has taken place at the time of training these AI models. There are certain similar issues when you talk about regulation; it is not a domain of law per se, but regulation is important from the perspective of how AI will seemingly integrate into various avenues.Coming to the latter question, as long as AI is used as a tool to complement human judgment, it should be an easy integration. The moment we as lawyers start getting AI systems to do our tasks, that can adversely affect even the practice of law. Human oversight and accountability need to be considered and if we talk of complete replacement, then I think there can be a threat to lawyers, although the entire practice cannot be replaced by AI systems. The reliance on how these AI systems are used in the practice, that would be the turning point..JC: Will AI lead to a redefinition of roles in the legal profession?.NN: Short answer, yes. With an introduction to anything new to any practice, be it law or any other practice, of course, there are changes in roles. The main question here would be to understand how we define the roles after the introduction of AI in the practice of law. The focus should not be on replacement, but rather on increasing efficiency. At present, integrating AI can free up a lot of time for lawyers to focus on strategic assignments, which can directly augment efficiency. When clients ask certain questions or demand predictions about what are the chances of achieving work in a certain period of time, with the advent of AI, we can get better with predictions. It shouldn't be the sole criteria and there has to be a mix of human judgment and AI tools. The roles will also vary depending on the size of a law firm. For sure, I think that there will be a change in roles. How we go about it and how we carefully integrate the use of AI are the critical questions..JC: What ethical considerations/guidelines do you think are essential for integrating AI into law?.NN: This is a good question, an extremely important one. Whenever we talk of integration of AI into any domain, and specifically when we are talking of the legal domain, there can be a lot of times when we need to question the systems that are developed. Whether the model is so transparent to understand how has it reached a particular judgment or whether the conclusion can be easily influenced.Second, what is universally debated are the biases that are inherently present in any AI system. How do we actually reduce these biases to make the systems more fair? Because in law and justice delivery, fairness is a critical aspect that we cannot overlook. Accountability for the decisions that are made either by AI systems or at least with the use of AI systems, is also important. Then there’s a lot of confidentiality clauses that come into picture. The majority of these have client details and data that are to be treated as extremely confidential information. So, at the time of development of various AI systems, it is important that at no given point of time, the confidentialities are violated. These are some of the important considerations to begin with before we start discussing integration of AI into law. With respect to these circumstances, there can be various soft law approaches that can be adopted by jurisdictions, so that we have a framework in place which governs how AI tools may be integrated more effectively and to increase efficiency in the legal domain..JC: Do you think unified, international regulation is important?.Rodney D Ryder (RDR): When we talk of AI regulation, we have seen a lot of jurisdictions adopting various policies and hard laws to regulate and govern the use of AI. How I look at it is whether there can be a harmonious framework.In my personal capacity, I would not advocate for unified regulation, because various countries and the jurisdictions that are currently using AI may be at different levels. So if we try to bind various countries together under a uniform code, that may hinder innovation to a large extent. It also becomes an important and critical factor to consider that the development that we are seeing in different jurisdictions in AI, can be extremely variable and the objects of each of these counties can be different. When you talk of international cooperation, I feel that it may be a very good idea to begin with the process of building sustainable and responsible AI systems. At this point of time, when AI is still growing, maybe it's not the right time to vouch for an international regulations. Maybe international cooperation with some soft law frameworks can be a good way to even let different jurisdictions achieve their individual goals with respect to innovation while safeguarding the building of sustainable and responsible AI systems..JC: What do you think are the implications of judges adopting AI to write judgments?.RDR: Using a tool is fine as long as I have full and complete oversight, because right now, the way in which AI is developing, it’s still in the initial stages. Although people are rightly talking about bias and risk, the main risk to my knowledge is that it doesn't have access to all the vast resources that are available, especially on Indian law. If judges get information from untrusted sources, how can that information be relied to wield the pen for justice? So there are two aspects of it. Have I done my work and whether my tool is wholly reliable.When AI is deployed at a larger scale than what we are seeing today, it will also lead to a lot more possibilities and a lot more new roles. As long as it is used as a tool, it's fine with oversight. So who does this oversight should mostly be a human and not an AI again.Even from a liability perspective, right now we haven't reached that stage where we can make machines liable. So if there is a goof up, who do you catch? You catch the human who's responsible. People's lives, people's fates, and justice, are at stake, and many a times for an eternity, because the precedent might be set and you can't have it on half-baked stuff.Right now, lwe are in what I would like to call the Stone Age of AI. It's very early days, so relying on it without reason, without cause is something that I wouldn't advise anyone to do. I would be very appalled if I come across people who are saying, 'This is written by AI'. I think that's grossly unfair. It shows that individual as a bit irresponsible and in the end it gives AI a bad name for no rhyme or reason. The poor thing can't handle that information. So to place that kind of burden and responsibility is just not fair..JC: What does AI integration mean for the pendency of cases? Would it expedite the process?.RDR: Definitely, without a doubt. Many clients ask us for a timeline after chargesheet is filed, the answer of which will depend on the lawyer's experience. Now AI tools can direct us to sections and legislation and suggest the average conviction term. Now to analyse large data sets dispassionately, across just plain statistics, is something that AI can do much better than Rodney or Nikhil can do. Our human brains cannot handle that much data. Plus, I would not be able to rattle facts from the Madras High Court suddenly, even if somebody puts a gun to my head. That's why I've been calling AI a wonderful tool.NN: It will also be great if we can identify some cases wherein deployment of AI can help us reduce litigation through litigation prediction. This is a tool that can predict the likelihood of Party A’s success versus Party B’s success in a case, and then whether such cases can be referred to mediation, arbitration or conciliation. So I think we will need to identify a lot of such cases and if AI helps us in this identification, it may help in reducing the current burden of pendency in the Indian courts..RDR: A simpler example is when clients ask us how long will it take for a patent to be granted or to get a trademark registered. We all give them a rough estimate. Now, an AI tool, looking at all the trademark applications filed this year will give you a precise time, because when we file a trademark application, the time and date stamp is logged in. Ultimately, we are concerned with outcomes and they're concerned with time and cost because time has an impact on cost as well. So I'm looking at basic data sets like “how much time?”. I'm wondering whether I should invoke the arbitration clause in this matter or not, how long will it take, what is my cost? But then there's no substitute at this stage for mixing that tool - which is about data and statistics - with someone's experience. A combination of both is ideal, instead of relying blindly on one thing or the other.Now why am I saying “one thing or the other”? In the era where we have such tools, it would be silly for me to just take a guess in the dark, but it would be equally silly for me to just take the tool and be lazy about it and not apply my mind..JC: What do you hope readers will take away from your book?.RDR: We've tried to give them a good foundation - introduction to all core topics - that's enough to start them on this journey. That's the idea of the book. We haven't tried to be overly ambitious. We have already started working on the third edition, so that kind of progress will be there.We've tried to straddle the worlds of both managers and people who are interested in AI, even non-lawyers. That's a hard task to do. We've sort of balanced it. The readers will give us feedback on whether we've managed to do that or not. This is a good, solid, robust introduction to the subject. It's one of those books that kickstarts a debate, and this is now an ongoing project.NN: This book is aimed at being a foundational guide, and it's for somebody who is venturing out wanting to understand what AI means, not just to the profession of law, but also how it interacts with various domains of law .This book can be also picked up by somebody who's not a lawyer and not a law student. It's important to understand that when you're making policies, you need resources to begin to start off to understand. So I think that is where this book fits in and serves the purpose..The book is slated to be launched on September 4 in the presence of Supreme Court judge Justice Surya Kant as the Chief Guest. Attorney General for India R Venkataramani and Senior Advocate Dr Abhishek M Singhvi are the Guests of Honour.