Artificial Intelligence: A Friend Or Foe For The Legal Industry?

Preeti Ahluwalia

26 May 2020 8:32 AM GMT

  • Artificial Intelligence: A Friend Or Foe For The Legal Industry?

    Artificial Intelligence (AI) is the most trending topic around the world these days,1 with multiple industries exploring their possibilities2 and its use cases emerging almost every day.3 The old and monotonous manual jobs requiring large numbers of skilled labours and workers are now can be done in just seconds with the help of computers.4 As technology advances it has paved a new...

    Artificial Intelligence (AI) is the most trending topic around the world these days,1 with multiple industries exploring their possibilities2 and its use cases emerging almost every day.3 The old and monotonous manual jobs requiring large numbers of skilled labours and workers are now can be done in just seconds with the help of computers.4 As technology advances it has paved a new concept commonly known as "Artificial Intelligence". Quickly after its inception AI has paved its way into different areas. Similarly the legal profession is also getting affected by the advancement of this technology and much truly has gone through an adoption of AI.5 But what is AI, and what is its relation to the practice and administration of law? The purpose of this article is to address those questions by providing an in-depth overview of AI and its use within law. This article aims to contribute to the body of knowledge of Law and technology.

    "AI is not a story of Human vs. Machine.6 Rather, it's human and machine vs. the problem."7

    History of AI?

    In 1956, John McCarthy commonly known as the 'Father of AI'8 articulated his notion of what is called AI in his first academic conference famously known as 'the Dartmouth Conference'.9 Later Alan Turing, a great mathematician10 introduced the idea of whether if it is possible to make machines have the same ability to think and learn by itself. He was able to put his theories and questions into actions by testing whether "machines can think and operate like human being"? After series of testing (which was later known as Turing Test) it turns out that it is possible to enable machines to think, learn and operate just like human beings.11 The Turing Test uses the pragmatic approach (something that is based on practical considerations, rather than theoretical ones) to be able to categorize if machines can respond as humans.12

    What is AI?

    AI is basically the area of study that defines the capabilities of machine learning just like humans and the ability to respond to certain behaviors.13 AI is designed using various algorithms that help the system to determine the probable answer, which will basically tell the system what to expect and work accordingly.14 All machines and software that use AI behave and interact like humans. Just the way humans need time and experience to grow physically and mentally, AI also needs time. It keeps developing with the experiences by storing data and applying it in future whenever necessary. AI technology has advanced and developed a large number of tools to solve the most difficult problems in computer science, i.e., logic, NLP (National Language Processing), control theory, search and optimization, probabilistic methods for uncertain reasoning and many more.15

    Currently, AI is being used in many areas, for instance, autonomous vehicle (drones and self-driven cars),16 Airport Security (Face detection),17 medical (diagnose of disease and design new drugs),18 playing games (chess),19 search engines and recommendations (Google search, YouTube, Amazon), online assistant (Siri, Alexa),20 Credit risk assessment,21 Photography (AI-powered cameras, portrait mode detecting depth in Android phones),22 Instagram application using machine learning to find and search function, target advertising, fight cyber bullying and removing of offensive comments on the posts,23 Uber cabs request for Demand-Supply, optimal pickup points, expected time of arrival for rides.24 In short, applying a brain into machines that allow them to take decisions by themselves is the literal meaning and purpose of AI.

    AI's presence, contribution and impact in the Legal Industry

    At its core, 'AI and law' are mainly concerned with applications of computer and mathematical techniques to make law more logical, convenient, beneficial, accessible, or probable.25 Theories of legal decision making, particularly argumentation model, is helpful for informed representation;26 the model of social organization has contributed to multi-agents systems; reasoning with legal cases has impacted to case-based reasoning;27 and the need to store and retrieve large amounts of textual data has resulted in contributions to conceptual information retrieval and intelligent databases.28 With that notion, one might trace the roots of similar ideas back to Gottfried Leibniz (a mathematician known for calculus and also a trained lawyer who researched how numerical formulas might improve the law) in the 1600s.29

    The terms AI & Law was first used in the 1970s by Buchanan and Headrick.30 The period includes Thorne McCarty's significant TAXMAN project,31 which was concerned with the modeling of the majority and minority arguments in a US Tax law case32 in the USA33 and Ronald Stamper explored LEGOL project that attempted to provide a formal model of the norms that administer the organization in the UK.34 In 1980, Carole Hafner's did some amazing work in this field, which included work on conceptual retrieval,35 and Anne Gardner's computational model is much discussed in jurisprudence, which helped differentiating between easy and hard case laws.36

    In 1987 a periodic conference, the International Conference on AI and Law (ICAIL), was formed, which is contended to be the foundation of AI and Law community. This conference was the main place for announcing and developing ideas within AI and Law,37 which later in 1991 led to the foundation of the International Association for Artificial Intelligence and Law (IAAIL).38 Examples in AI and Law include Valente's functional ontology (the term 'ontology' refers to the study of conceptualisation) and since then Legal ontologies have become the topic of regular workshops at AI and Law conferences.39

    In 2018, a study depicted that LawGeex, which is an AI contract review platform, tried it on human Lawyers and surprisingly the system achieved an accuracy rate at 94% surfacing risks in Non-Disclosure Agreements (NDAs) in comparison to experienced human lawyers who average out at an 85% accuracy for the same task. The legal AI system took 26 seconds to complete the review whereas the human lawyers took an average of 92 minutes.40 The companies are using AI on their daily legal works, including drafting and assessing contracts, mining documents in discovery, answering regular questions or examining data to predict outcomes.41 Legal AI means lawyers are gradually determined on the current realities and future of the law, rather than just the examples of the past.42 Despite defeating lawyers at smaller tasks, experienced and educated lawyers are still required as trusted advisers in important deals. The goalmouth is to use AI as well as human brain to ensure that research is done more precisely and more consistently than a human or technology alone.43

    Pros and Cons of AI Technology in the Legal Industry

    Half of the world is trying to convince others that AI is an 'enemy' for the legal industry whereas few have already proclaimed it as a 'friendly' tool. Therefore, the sole objective of this part is to throw some light on both of these perspectives in the form of pros and cons.

    A. Pros:

    i. Due diligence reviews – 'Due diligence' is one of the main responsibilities that lawyers perform on behalf of their clients viz, the validation of facts and figures, carefully assessing risks, and structuring an agreement. This process is needed for advising clients about all the possible options they have, and what actions they should take. Generally lawyers conduct a comprehensive analysis (from collecting documents to examining each text for key data points) for meaningful outcomes. As such, lawyers are also prone to mistakes and inaccuracy when analyzing spot checks. On the other hand, technological advancement have changed the way many MNCs conduct due diligence. With AI technology, the work can remarkably be done faster, while guaranteeing greater accuracy in their reporting and saving lawyers time and energy that can be used more productively elsewhere. There are numerous tools (example, iManage Diligence Engine, Kira Systems, LawGeex, Legal Robot, Casetext) that can automate this process using AI technology.44

    ii. Legal Research & Time Saving – Another evident advantages of AI is legal research and time consuming quality. AI driven legal research has a great potential in reducing the time required to prepare quality research. In today's legal panorama, clients expect to get the best opinion possible at a reasonable price. Companies and law firms that adopt AI legal research software are being proactive at guaranteeing their clients walk away knowing they have received the best legal opinion without paying extreme legal fees.45

    iii. Prepare and create contracts – Another benefit of AI devices is that they can create contracts by utilizing effective tools. These devices act as a self-service tool for clients, for example, Contract Express, which is contract automation software that helps the client to create and produce agreement they require, enter few terms and conditions, and the structure will deliver a standard shape understanding prepared to go.46

    iv. Error detection and reduction – With AI coding, it is now possible to look at the organization records like posts, reports and unstructured information and detect unpleasant conduct before it can rise to the surface.47

    There are numerous other advantages of AI, for instance, it improves organizational and logical structure, which help lawyers recognize gaps and holes in their reports and even in their legal research. Further the most significant benefit of AI is that it reduces lawyers stress in many ways, for example, research, documentation, editing, assembling can be tedious and time-consuming task.48 With the help of AI all that hard work can lessen the pressure and this in turn, can enhance the creative analysis of lawyers. Further, prediction technology has the capability to analyse data to help it make predictions about the outcomes of the cases, legal analytics used to determine trends and patterns from precedent laws, judges' history and win/loss rates. AI tools help in providing insight into the IP portfolios, and most importantly electronic billing, which help lawyers in preparing the invoices as per the work done by them.49

    B. Cons:

    i. Taking away of jobs – As said by Sundar Pichai, the CEO of Google, "In the next decade, Google will shift from mobile-first to AI-first" (that's why Google Assistant is designed, which allows people to have a natural conversation between you and Google. AI could replace the jobs of office clerks, administrators, customer service representatives, analysts, and underwriters in the next decade.50 As per the data, it is estimated that nearly around 100,000 legal-related jobs will be automated by 2036 because of AI.51 There are numbers of AI devices that are replacing jobs in the legal sector; for example, an Australian legal tech company has built an aprototype virtual lawyer using Amazon's Alexa that can create legal documents much faster than an actual lawyer.52 Major law firms are using AI machine learning as machines can look over many thousands of reports and locate the applicable data for a case much faster than a human can.53

    ii. Things can possibly go wrong – If untrained professionals at the lower level are working and using the technology then there are chances that many errors and mistakes can occur.

    iii. Vulnerable to attack – Experts are concerned about the cyber security, which may not be as impactful as depicted.

    iv. Economic problem – AI being a handy tool brings a lot of expenses too. From its formation, repairing, to programming checks, it requires tremendous expenditures, as they are exceptionally complex machines. On account of serious breakdowns, it may require tremendous time and cost to retrieve the lost codes and reinstating the entire framework.54

    How the various pros can overshadow the cons?

    So far we have discussed about the various aspects of AI on the Legal Profession including its pros and cons. Now we shall focus on how the various pros can overcome the cons as discussed in the earlier part of the article, which AI is deemed to have on the legal industry.

    Nowadays, there is a burning question as well as the biggest fear among the lawyers that whether the introduction of AI technology in the legal domain would replace the lawyers? And in turn, use of AI based solutions and platforms would increase the productivity and efficiency of the legal domain. The legal industry has seen much of the technological advancement in terms of contract analysis, trademark search software, legal research software and much more. However, none of these AI based solutions plan to take a lawyer's job, rather they increase the authencity, accuracy of research and analysis in many different ways. In an article, it can be seen that the legal domain is likely to be effected by just 3.25%, which is among the lowest in the list.55 AI will not replace Lawyers, but in the upcoming years, lawyers who will use this technology will definitely replace those who don't use it. Now answering the question about trained lawyers losing their jobs. In a nutshell, if the research and documentation process is been done by the machines, the trained professionals can focus on the other important tasks for instance, preparing for better arguments, case reports; maintain good client relations, etc.56

    Another issue, which the people believe to have on the use of AI is that if it is used by untrained professionals. As per the data, a study shows that the technologies margin of error would be reduced from 3% to +-0.01%.57 The last con on, which the people argue in against of using AI is its tremendous cost that this tech brings along as an economic burden. With the saving of money in terms of the salary paid to the paralegals and researchers can be utilized and revenue generation can increase.

    Above all, AI tends to save a lot of time of lawyer's. It is cost effective, which enables the law firm to generate more revenue. The AI helps the lawyer to maintain good client relations. If the work, which takes upto 10-12 hours to complete, it can be done within few seconds with the adoption of AI technology.

    Legal Framework in India

    India, being a developing nation, has a considerably huge population and people outstripping in the field of Law and Technology. The Ministry of Industry and Commerce in India, has constituted 18-member task force titled, "Task Force on AI on India's Economic Transformation"58 to explore all the possible outcomes and in 2018 the task force has given a detailed report providing steps to be taken with regards to the formulation of an comprehensive policy on AI in India. Task Force has also recommended to the Department of Industrial Policy and Promotion (DIPP)59 to set up and fund an Inter-Ministerial national AI Mission, for a period of 5 years, with funding of around 1200 crores, to act as a nodal agency to co-ordinate all AI related activities in India.60 Setting up of data banks and ombudsman will be the duty of Ministry of Electronics and Information Technology (MeitY). While DIPP will look after the formulation and implementation of the regulations related to ownership, privacy and security.61 The report also proposes to Bureau of Indian Standards to ensure that India participates and implements all the required norms with regard to AI systems.62 The report also suggests that DIPP and MeitY should work towards the enactment of policies namely, policy with data, which deals with ownership, sharing rights and usage data and tax incentives for income from AI applications. Ministry of Human Resource Development and Ministry of Skill Development and Entrepreneurship should look after education curriculum to develop adequate human resources with required skill sets to meet the growing demands of professionals who can handle AI systems.63 Finally the report proposes for inter-ministerial collaborations to ensure that India actively participates in all the meetings and conferences on AI conducted at international forums.64 There are few conferences that are taking place in India, like, The Machine Conference 2020, Data Science Congress 2020, The Rising 2020, Gartner Data & Analytics Summit 2020.65 India has taken many initiatives for the formulation and implementation of AI. Though, the liability and punishment part of this technology is lacking, as there is no legal personality as of now. There are number of theories suggesting on how AI can be brought under criminal and civil law: for instance, Natural consequence theory where the user will be held accountable for their negligent rational state, innocent agent theory where the programmer will be held liable and direct approach theory where independent AI entitiy will be responsible for its own act. Tort of negligence would require establishing that AI owned a duty to take care, a duty towards the plaintiff, and damage was caused to the plaintiff due to breach of such duty by AI. Since machines have been treated as consumer products therefore, strict liability rule has been assigned to the manufacturers. At present, AI entity is not held liable due to lack of a legal personality.66

    The goal of this article was to provide a realistic and demystified view of AI and law. The purpose of applying AI based legal solutions and deep machine learning methods is to increase the overall performance of legal systems. From the aforementioned discussions, it can be concluded that if the lawyers and law firms use AI technology, it is going to improve the efficiency in research process, the use of software can pacify the process of document review, a lot of resources will be saved in form of time and money. The world is changing with increase in technological advancements. AI is leading in multiple sectors by automating jobs of legal professionals at a certain level, but this won't up take their jobs instead it will lead them to various different roles. It is difficult to make a statement about how the whole process of AI in the legal industry will look after a couple of decades but as of now it can be said that the adoption of AI in the legal industry is worth a shot. The innovation of every technology comes with some issues, but that doesn't mean it's useless. It is very important to comprehend the new technology in well timed and suitable regulatory framework is established so that the entire world does not miss out from a gigantic opportunity.

    Views Are Personal Only


    [1] Sonia K. Katyal, Private Accountability in the Age of Artificial Intelligence, 66 UCLA L. REV. 54 (2019)

    [2] Michael Cross, Top 5 sectors using artificial intelligence, raconteur (Dec. 15, 2015),

    [3] Gaurav Batra, Andrea Queirolo, and Nick Santhanam, Artificial Intelligence: The Time to act is now, McKinsey & Company (Jan, 2018),

    [4] Dom Galeon, An AI Completed 360,000 Hours of Finance Work in Just Seconds, Futurism (Mar. 8, 2017),

    [5] Nick Ismail, Artificial intelligence in the legal industry: Adoption and strategy – Part-1, Information Age (Aug. 6, 2018),

    [6] Aurecon, Artificial intelligence vs human intelligence – a competition or collaboration?, Aurecon

    [7] Vasanth Kumar, Artificial Intelligence vs Human Intelligence, great learning blog (Mar. 19, 2020),

    [8] V. Rajaraman, John McCarthy – Father of Artificial Intelligence, 198 (Mar, 2014),

    [9] See in John McCarthy et al., A Proposal For the Dartmouth Summer Research Project on Artificial Intelligence August 31, 1955, AI Magazine, 2006 at 12

    [10] James Grime, What did Turing do for us?, NRICH (Feb, 2012),

    [11] B.J. Copeland, Alan Turing and the beginning of Artificial Intelligence, Encyclopædia Britannica (Mar. 24, 2020),

    [12] Dave Roos, How Alan Turing and his test became AI Legend, howstuffworks (May. 22, 2019),

    [13] Jean-Paul Haton, A brief introduction to artificial intelligence, Science direct assests (2006),

    [14] Fahrettin Filiz, Artificial Intelligence Algorithms, towards data science (Mar. 31, 2017),

    [15] Zoublin Ghahramani, Uber AI IN 2019: Advancing Mobility with Aritificial Intelligence, Uber Engineering (Dec. 18, 2019),

    [16] Noah Rue, AI, Self-Driving Cars, and Lyft/Uber: Helpful or Harmful?, M (Dec. 21, 2018),

    [17] Millicent Abadicio, Facial Recognition at Airport – Current Applications, Emerj: The AI Research and Advisory Compliance (Nov. 22, 2019),

    [18] Nicola Davis, AI equal with human experts in medicial diagnosis, study finds, The Guardian (Sep. 24, 2019, 23:30 BST),


    [20] R.L. Adams, 10 Powerful Examples of Artificial Intelligence in use today, Forbes (Jan. 10, 2017, 08:32 AM),

    [21] Nadège Grennepois and Erwan Robin, Explain Artificial Intellgence for Credit-Risk Management, Deloitte(July,2019),

    [22] Sam Byford, How AI is changing photography, The Verge (Jan. 31, 2019, 8:00 AM EST),

    [23] Staff Reporter, How Instagram uses Artificial Intelligence in 2019, The Science Times (Nov. 29, 2019, 06:27 PM EST),

    [24] Jamal Robinson, How Uber organizes around Machine Learning, Medium (Feb. 4, 2019),

    [25] See V. Vaissnave and P.Deepalakshmi, An Aritifical Intelligence based Analysis in Legal Domain, 9 International Journal of Innovative Technology and Exploring Engineering (IJITEE) 1046 (2019)

    [26] See Marcello Di Bello and Bart Verheij, Evidence & decision making in the law: theoretical, computational and empirical approaches, Artificial Intelligence and law 28, 1-5 (2020),

    [27] See David Fitoussi and Moshe Tennenholtz, Choosing social laws for multi-agent systems: Minimality and simplicity, Aritificial intelligence 119, 61-101 (2000),

    [28] Vaissnave, supra 1046

    [29] See Giovanni Sartor, A Treatise of Legal Philosophy and General Jurisprudence, 5: Legal Reasoning 389-90 (Enrico Pattaro ed., Springer 2005)

    [30] See Bruce G. Buchanan and Thomas E. Headrick, Some Speculation about Artificial Intelligence and Legal Reasoning, 23 Stanford Law Review 40 (Nov, 1970)

    [31] L. Thorne McCarty, Reflections on" Taxman: An Experiment in Artificial Intelligence and Legal Reasoning 90 Harvard Law Review 837 (Mar,1977)

    [32] Eisner v. Macomber, 252 U.S. 189 (1920)

    [33] Blazej Kuzniacki, The Marriage of Artificial Intelligence and Tax Law: Past, Present, and Future, The Amsterdam Centre for Tax Law (University of Amsterdam) (Feb. 11, 2019),

    [34] See Ronald Stamper, Biography, British Computer Scientist,

    [35] Carole D. Hafner, Conceptual organization of case law knowledge bases in ICAIL 87 (1987)

    [36] Edwina L. Rissland et al., AI and Law: A fruitful synergy, 150 Artificial Intelligence 1 (2003)

    [37] ICAIL 2015—First Call for Papers, INT'L ASS'N FOR ARTIFICIAL INTELLIGENCE (Sept. 10, 2014), []

    [38] See

    [39] See Pepijn R. S. Visser and Trevor J. M. Bench-Capon, A Comparison of Four Ontologies for the Design of Legal Knowledge System, 6 Artificial Intelligence 27 (1998)

    [40] LawGeex, Artificial Intelligence More Accurate Than Lawyers for Reviewing Contracts, New Study Reveals, Cision PR Newswire (Feb. 28, 2018),

    [41] Prelaw, What Do Lawyers Do?, NALP,


    [42] Thomson Reuters, Ready or Not: Artificial Intelligence and Corporate Legal Departments, Thomson Reuters,

    [43] H. James Wilson and Paul R. Daugherty, Collaborative Intelligence: Humans and AI are Joining Forces, Harvard Business Review (Jul-Aug, 2018),

    [44] iManage, How Today's Legal Professionals are Using AI for Due Diligence, iManage (Nov. 20, 2018),

    [45] See Tara Vasdani, AI legal research tools save money, time and judges' enmity, Canadian Lawyer (Mar. 25, 2019),

    [46] Thomson Reuters, Contract Express, Thomson Reuters,

    [47] J.K.C. Kingston, Artificial Intellgence and Legal Liability (2018), arxiv,

    [48] See David Grossi et al., Organizational structure and responsibility: an analysis in a dynamic logic of organized collective agency, ACM Digital Library (Jan, 2007),

    [49] ROSS Intelligence, AI for the practice of law: An introduction, Ross Intellgence bloq (Aug. 8, 2018),

    [50] Harshith Mallya, How Google is going from mobile first to AI first while competition heats up, your story (Oct. 5, 2016),

    [51] Bernard Marr, How AI and Machine Learning are Transforming Law Firms and the legal Sector, Forbes (May. 23, 2018, 12:29 AM),

    [52] Byron Connolly, Artificial Intelligence putting junior lawyer jobs at risk, CIO (Jan. 15, 2019, 16:57 IST),

    [53] Alison DeNisco Rayome, Why human-AI collaboration will dominate the future of work, Tech Republic (June. 4, 2018, 4:00 AM),

    [54] Irving Wladawsky-Berger, The impact of Artificial Intellgence on the World Economy, The Wall Street Journal (Nov. 16, 2018, 3:09 PM),

    [55] See Carl Benedikt Frey and Michael A. Osborne, The future of employment: How susceptible are jobs to computerization?, 114 Technological Forecasting and Social Change 254 (2017)

    [56] Steve Lohr, A.I. is doing legal work. But it won't replace lawyers, yet. The New York Times (Mar. 19, 2017),

    [57] Spain's News, Artificial intelligence has a margin of error of 3%; humans, 5% "| Innovation, Spain News,

    [58] See

    [59] Viksit Gaur and Akriti Gaur, On AI, various government agencies have conflicting and confusing strategies, Indian express (Oct. 11, 2019, 11:29 AM),

    [60] Department for Promotion of Industry and Internal Trade, Report of Task Force on AI, Department for Promotion of Industry and Internal Trade (2018),

    [61] Surabhi Agarwal and Yogima Seth Sharma, Govt sets up panel to resolve MeitY and Niti fight over AI, The Economic Times (Oct. 23, 2019, 09:00 AM),

    [62] Press Information Bureau, Implementation of BIS Act, 2016, Ministry of Consumer Affairs, Foods & Public Distribution (July. 16, 2019, 4:01 PM),

    [63] MHRD, National News, News Service Division (Sep. 12, 2019, 7:23 PM),

    [64] Department for Promotion of Industry and Internal Trade, Report of Task Force on AI, Department for Promotion of Industry and Internal Trade (2018),

    [65] Anu Thomas, TOP 6 MUST-ATTEND AI & ML CONFERENCES IN INDIA FOR 2020, CIM (Feb. 10, 2020),

    [66] Aprajita Tyagi, E-Commerce from push/Impulsive to Intelligent, 5 Journal of Legal Studies and Research 78 (2019)

    Next Story