PROFFESIONAL ETHICS

CENTRAL UNIVERSITY OF SOUTH BIHAR

School of law and governance

AI AND LEGAL OUTSOURCING IN INDIA: OPPORTUNITIES

AND CHALLENGES

SUBMITTED TO: Dr. Ashok Kumar

(ASSISTANT PROFFESOR)
SUBMITTED BY: Mishra Anshal Kashyap
(cusb1913125063)
10th semester, sec-(A)
 ACKNOWLEDGMENT
It is a great pleasure to express deep sense of thanks and gratitude to my course instructor and
guide Dr. Ashok Kumar. His dedication and keen interest above all and his overwhelming
attitude to help his students had been solely and mainly responsible for completing my work.
His scholarly and timely advice, meticulous scrutiny in such difficult and unprecedented
times, and his logical approach has helped me to a very great extent to accomplish my
assignment in an excellent manner.
I owe a deep sense of gratitude to Professor S.P. Srivastava, Head & Dean of School of Law
and Governance, for providing us with such a learning assignment that helped us to properly
understand. It is my privilege to thank my parents for their constant encouragement
throughout the period and providing me with needful resources at such stressed and
unprecedented times.
Table of Contents

 Introduction

 Understanding the technologies

 Artificial Narrow Intelligence and Artificial General Intelligence
 Machine learning
 Deep learning
 Data science

 Legal tasks which can be automated

 Legal research
 Reviewing contracts
 Due diligence
 Contract analytics
 Litigation prediction
 Assisting in registration of intellectual properties
 AI chatbots

 Current projects working on automating legal industry

 Smith.AI
 Casetext
 Diligen

 Artificial intelligence in legal process outsourcing

 The need for ai in legal sector

 Challenges of implementing ai in legal process

 Conclusion
Introduction
In today’s day and age, we are having very active artificial intelligence softwares in our
pockets. They are right there in the devices which we use very frequently, like our mobile
phones, tablets and laptops. These softwares know about our preferences and needs,
especially regarding the content we consume on a daily basis. According to a report by
Mckinsey Global Institute, Artificial Intelligence (AI) is going to create $13 trillion value in
various industries. This clearly indicates that every part of life will have some tasks which
will be automated by artificial intelligence.

One such part will surely be the legal industry.

All people, at least once in their lives, have to interact with the legal system. They might have
to either register some intellectual property they created, their own company, or their
marriage or might even have to fight a legal suit. Simply put, most individuals, in some or the
other case, will have to interact with the legal system.

Talking about the pain points in the legal industry, especially in India, it is that all the
matters, as well as procedures in the legal industries, are very time-consuming. The reason
behind the same is that mostly all the tasks that are to be done by legal professionals such as
legal research, drafting and vetting of contracts, due diligence, etc., take up a lot of time. At
times, some of these tasks become quite monotonous in nature, which is why several
independent practitioners, as well as small and big law firms, hire numerous junior-level
lawyers. The main tasks which are done by these lawyers include legal research, drafting the
first level of contracts, etc. The seniors, most of the time, delegate these tasks to the junior-
level lawyers and they themselves handle the tasks like providing consultation to the clients,
negotiating contracts on behalf of their clients, representing their clients in courts and
reviewing and vetting the legal documents drafted by the junior-level lawyers.

As Artificial Intelligence is a very promising prospect for all industries, its application in the
legal industry can create a major impact on the lives of people.
In this article, we will try to understand how Artificial Intelligence, machine learning, deep
learning and data science work and which are the tasks in the legal industry that can be done
effectively by these technologies, improving the lives of the lawyers as well as their clients.

Understanding the technologies

Artificial Narrow Intelligence and Artificial General Intelligence

To determine which tasks can be replaced by technology, the first step is to understand how
these technologies work and what their basic operating principles are. So let’s first, one by
one, understand how these technologies work.

When it comes to artificial intelligence, it is a technology that allows computers or machines

to think and make decisions. At their very core, computers are still dependent on humans, and
they need a very large amount of data to create a pattern and then perform the tasks based on
the information that is fed to them in the form of data previously. The most common
misconception that people have about Artificial Intelligence is that it can automate all the
tasks, but the reality is a bit different. There are two types of Artificial Intelligence, namely: 1

1. Artificial Narrow Intelligence, and

2. Artificial General Intelligence.

The first one is a much more feasible technology as of now because, as per the growth of
technology we have achieved, we can only train a machine to do a particular task using
Artificial Intelligence such as speech recognition, language translation, etc. To achieve
Artificial General Intelligence, we would have to train the machine to think in a similar way
humans are able to do and do all the tasks, and that is not possible as of now because we have
not yet found a way to figure out the working of a human mind.2

1
Mudit gupta,`Ai and how it can change legal industry’(ipleaders, 13 feb 2023) https://blog.ipleaders.in/artificial-
intelligence-and-how-it-can-transform-the-legal-industry/ accessed 30 April 2024
2
Ibid1
Machine learning

Now, let’s understand what machine learning is. Machine learning is a method of achieving
AI. It is a way of teaching computers to learn from data without being explicitly
programmed. In machine learning, algorithms are used to identify patterns in data and make
predictions or decisions without being explicitly programmed to perform the task. It is
important to understand that achieving 100% accuracy from a machine learning algorithm is
not realistic. If an algorithm is able to achieve near 95% accuracy, then that machine learning
algorithm is good enough to deploy. But this much accuracy should be achieved on a
significant amount of testing data. This technology basically works on the concept of A to B
mapping in which the machine is first fed with the data of both the inputs and the outputs,
and at the testing stage it is asked to complete the task by inputting A for which it gives the
outputs, which are B. For example, if a machine learning algorithm is fed data on which
emails are spam and which are not, we must first provide input of both types of emails and
tell it which is the spam mail and which is not, and then at the testing stage, we will enter a
mail and the machine will output if it is a spam mail or not based on the data that was given
to the algorithm.3

Deep learning

Now that we have understood what artificial intelligence and machine learning are let’s
understand what deep learning is. It is a type of technology that is used in machine learning.
It is a technology for building and training large neural networks, which enables the machine
to accomplish the task for which such a network is built. This technology was built with
inspiration from how the human brain works, but their working processes are completely
unrelated.

3
Mathew Stepka, `Law bots’ (business law today, 21 feb 2022)<https://businesslawtoday.org/2022/02/how-ai-
is-reshaping-legal-profession/> accessed 30 April 2024
Data science

Now let’s briefly understand what data science is. Data science is a field that uses scientific
methods, processes, algorithms, and systems to extract knowledge and insights from
structured and unstructured data. It involves the collection, cleaning, analysis, and
interpretation of data, as well as the use of various techniques such as statistical modelling,
machine learning, and visualisation to extract meaningful insights. This, in turn, helps in
making informed decisions.

Legal tasks which can be automated

Now that we have understood how these technologies work at their root level and what kind
of tasks we can automate in order to decrease the time consumed, which in turn can improve
the situation of humankind, let’s try to understand which are the tasks in the legal industry
that can be replaced by these technologies.

Legal research

In the legal profession, precedents are of utmost importance. In law offices, lawyers and
interns are busy researching precedents where similar issues were discussed in superior
courts. The decisions given and the interpretations of the senior-level judiciary can help them
prove their point in a much easier and more effective way. This can save a lot of time, and the
problem of delayed delivery of justice can be solved in much less time.

The legal research is basically done by providing keywords to the software and then
analysing the various search results it provides to reach a similar conclusion. This task in the
legal industry can be automated, saving a lot of time and energy, which in turn can help
humankind.

To solve this problem, we will have to make the computer learn from the data from the
“Keywords” and the desired results, and then at the trials, we will have to input the problem,
and it will give us the results based on the previously fed data.

Although these results are fairly accurate, they cannot be completely relied upon, and as law
is a profession of trust and conviction, these results will have to be reviewed to some extent.
Reviewing contracts

In today’s world, where there is a boom of entrepreneurship and people are eager to start new
companies that provide very feasible solutions to real-world problems, the founders most of
the time struggle to handle the legal side of their operations, in which they have to contract
with many other stakeholders on a regular basis through written contracts. In real life, the
complete life cycle of signing a contract can sometimes become very tedious as it requires a
lot of negotiation and vetting of the legal documents. This time duration of a contract life
cycle can be reduced, which can make this whole procedure very much streamlined and can
save a lot of time for lawyers so that they can help more and more clients in a short amount of
time.4

The main thing that can be done to automate this part of the legal industry is to create AI
systems that can automatically absorb proposed contracts, examine them properly by
adopting natural language processing (NLP) technology, and decide which aspects of the
contract are fine and which parts may pose a problem.

Due diligence

The main task when a corporate transaction takes place is to work on due diligence. In this
task, law offices work on behalf of their clients to check the past history and the paper trail of
the other party they are dealing with. It involves checking for certain things, such as any past
litigation, and if they are found, they have to be discussed, and then any further decision is
made to take the deal forward.

This task can be very well managed with the use of technology by automating it and
standardising what needs to be checked and where it needs to be checked.

Contract analytics

One study estimates that poor processes for contract management cost businesses 9% of their
annual revenue. This clearly indicates that the lack of an effective mechanism for contract
analysis is a big enough problem that it is causing hindrances to businesses in scaling their
operations. These issues can be very effectively solved by the introduction of Artificial
Intelligence.

AI software can significantly reduce the amount of human power required to update and edit
contracts to reflect the specific agreement between two parties, which, in turn, will also lead
to effective time management. Every issue regarding the updating and novation of an

4
Ibid2
agreement can be very well solved using machine learning and language analytics by using
the A-to-B mapping systems.5

Litigation prediction

Most of the time, when the lawyers are interacting with the clients regarding a litigation suit,
one of the prominent questions the clients have is regarding the chances of winning the case.
Lawyers make predictions about the outcome of litigation based on their experience and
understanding of human psychology. Sometimes people just file a suit for the sake of it.
These suits can be reduced if we have a mechanism that can predict the outcome using the
judgments given in similar cases previously. This can significantly reduce the number of
cases filed and, hence, reduce the burden on the judiciary.

Assisting in registration of intellectual properties

Nowadays, people are innovating new things and starting new ventures. To secure their
innovation and to create an innovative business, it is very important for the inventors and
founders to secure their innovation by way of patents, trademarks, and copyrights. This task
mostly requires a lot of past checking and a lot of paper trail. This system can be very
efficiently automated, and a lot of human energy and time can be saved by feeding the
machine the data about all the granted intellectual property and, on that basis, letting the
machine decide whether it can be granted or not.

AI chatbots

Sometimes, lawyers are too busy representing their clients in court and don’t have much time
to sit with their clients and solve every single issue. This problem of their clients can be
solved by the AI chatbot, which can be an integration of 2-3 projects such as litigation
prediction, assisting in the registration of intellectual properties discussed earlier in this
article, and on the basis of that, the system can generate customised answers for some
specific questions that are common for most of the clients and help them get answers for the
questions, which can help them calm their nerves because sometimes they can be quite
nervous while handling legal suits.6

Current projects working on automating legal industry

Now that we have learned about the working and application of Artificial Intelligence and
other technologies and how they can automate tasks in the legal industry let’s take a look at
some of the projects that are solving these problems with the help of these technologies.
5
Rob Toews, `Ai will transform law’(Forbes, 19 dec 2019)<
https://www.forbes.com/sites/robtoews/2019/12/19/ai-will-transform-the-field-of-law/?sh=265e3cd67f01>
accessed 30 April 2024
6
Ibid3
Smith.AI

This is a startup that has tried to assist law offices by providing them with a “virtual
receptionist” with the help of an AI chatbot that helps in logging and transferring the calls
that a law office receives on a daily basis. It is an outsourcing service that provides services
to lawyers in their region.

Case text

This software works towards providing automated legal research to lawyers in just one click.
This allows them to spend more time preparing their arguments, counselling their clients, and
performing other tasks that only human intelligence can perform.

Diligence

This software assists law offices in doing due diligence by automating the process to some
extent. This software employs machine learning technology to examine contracts for specific
clauses, provisions, or changes, as well as to generate a quick and convenient summary of the
clauses in the documents.

Artificial intelligence in legal process outsourcing

The legal industry is fabled for accepting changes at a glacial pace. The lack of technical
knowledge and expertise as well as the fear of job losses have fuelled the reluctance to
adopt automation in the legal industry. Studies show that more than 80% of law firms fear
that technology will replace human resources such as lawyers and paralegals.
However, artificial intelligence (AI) has benefitted both people and processes in every
sector, and the legal services industry is no exception. The impact of AI on the legal
profession is expected to be huge in managing the entire legal process efficiently. The
legal AI software market size is anticipated to grow from $317 million in 2019 to $1,236
million by 2024, at a CAGR of 31.3%. 7

The need for AI in the legal sector

Implementing AI and AI-related business models is set to transform the legal sector in
areas such as:

7
Michael B Sonsteng, `legal outsourcing’(infosysbpm)< https://www.infosysbpm.com/blogs/legal-process-
outsourcing/artificial-intelligence-in-legal-process-outsourcing.html> Accessed 30 Apr. 24
  Software equipped with AI capabilities, such as lawyer bots, will automate lower-
level repetitive manual tasks so that attorneys can focus on other complex jobs.
 AI in the legal sector will reduce operational costs and increase speed, efficiency,
productivity, and accuracy. According to a research conducted by one of the world’s
leading management consulting firms, AI has the power to automate up to 25% of a
lawyer’s time.
 Intelligent machines will help gather and analyse data to predict the outcome of
litigations.
 An AI-led approach will save time, enabling lawyers to serve more clients and
prioritise cases.
 While automation seems to threaten jobs, research shows otherwise. AI will
generate new jobs and skill sets in the legal field such as legal engineers, legal data
scientists and architects, legal strategy consultants, etc.
 Automation in the legal firms will help achieve faster response times and more
accurate responses to the client.
 Clients now demand transparency in billings of the legal counsel. AI technology
helps analyse legal bills, enhance transparency with the clients, and increase
efficiency, which will result in law firms gaining more business. 8

Challenges of implementing AI in legal processes

Despite the benefits of AI, it is crucial to understand why the legal industry is slow to
adopt technological advancements. Let’s look at the underlying reasons:

 Cybersecurity needs to be addressed with robust systems and secure tech, which
most law firms presently lack.

 Integrating AI into existing systems requires specialists. This ensures that the AI
capabilities are top notch and the feedback loop is built to update existing systems
continuously.

 Law firms are challenged in terms of hiring resources such as experts who are
proficient in both AI and legal services.

 Since legal AI is in a growing phase, vendors of legal AI technologies may find it

challenging to give numerous case studies to potential new clients.

 AI systems need to remain updated about the daily changes in the legal regulations
to prevent giving any wrong predictions.

8
Michael B Sonsteng, `legal outsourcing’(infosysbpm)< https://www.infosysbpm.com/blogs/legal-process-
outsourcing/artificial-intelligence-in-legal-process-outsourcing.html> Accessed 30 Apr. 24
However, legal industry trends say that over 80% of senior lawyers deem AI and machine
learning (ML) crucial for their success. 9

Conclusion

Artificial Intelligence is a reality today. It cannot replace human beings but can assist them in
various tasks that they do on a daily basis. All tasks in which decision-making depends upon
some kind of pattern recognition and which are quantifiable in any manner, meaning
Artificial Intelligence can replace some of the tasks which are dependent on the Intelligent
Quotient (IQ), but today’s standards of technology cannot replace the tasks which require
Emotional Quotient (EQ). The reason behind this is very simply explained by a quote by
Yuval Nova Harari: -

“Humans think in stories rather than in facts, numbers, or equations, and the simpler the
story, the better.”

This uniqueness in our thought patterns cannot be accommodated by current technology, and
law as a profession deal with people’s emotions as well as their psychology. This means that
most tasks cannot be replaced by current technology, but current technology can help us
develop some tools that can help us deliver value to our clients and create more time value
for ourselves and our clients.

9
Ibid4
 References
1. https://www.forbes.com/sites/robtoews/2019/12/19/ai-will-transform-
the-field-of-law/?sh=56045c8d7f01
2. https://legal.thomsonreuters.com/en/insights/articles/ai-and-its-
impact-on-legal-technology
3. https://businesslawtoday.org/2022/02/how-ai-is-reshaping-legal-
profession/
4. https://www.barandbench.com/columns/artificial-intelligence-and-
legal-profession-an-intelligent-way-ahead
5. https://jolt.law.harvard.edu/digest/a-primer-on-using-artificial-
intelligence-in-the-legal-profession
6. https://www.infosysbpm.com/blogs/legal-process-outsourcing/
artificial-intelligence-in-legal-process-outsourcing.html
7. https://businesslawtoday.org/2022/02/how-ai-is-reshaping-legal-
profession/