Content uploaded by Divya Saxena
Author content
All content in this area was uploaded by Divya Saxena on Jul 01, 2024
Content may be subject to copyright.
ANVESAK
ISSN : 0378 – 4568 UGC Care Group 1 Journal
Vol. 53, No.2 July – December 2023 57
ARTIFICIAL INTELLIGENCE: AN INTERPLAY INTO THE MODERN LEGAL
JURISPRUDENCE
Ms. Divya Saxena
Assistant Professor, Bharati Vidyapeeth (Deemed to be University)
New Law College, Pune
Ms. Niha Khan
Assistant Professor, Bharati Vidyapeeth (Deemed to be University)
New Law College, Pune
ABSTRACT
The emergence of Artificial Intelligence (AI) has raised various ethical and legal concerns due
to its potential impact on society. AI has become a game-changer in different sectors, and its
potential for revolutionizing industries is enormous. However, its development and deployment
must consider ethical principles and comply with existing laws. This paper aims to provide an
overview of the laws and ethics surrounding AI, highlighting the challenges that arise as AI
advances. It covers privacy, transparency, bias, accountability, and responsibility. The paper
concludes by proposing measures to ensure that the development and deployment of AI comply
with ethical principles and existing laws.
Keywords: Artificial Intelligence, Law, ethical Concerns, Supreme Court, Legislation.
INTRODUCTION
Typically, the classification, accuracy, and reliability were determined by reviewing documents,
which have always been considered the gold standard in the market. This gold standard bar is
increasing in a world where technology facilitates searches, analyses, reviews, and
documentation for legal proceedings. Similarly, In Judgement making, humans are also expected
to make judgments with computer precision and consistency across large data sets. Seeking
computer-like consistency from people and human-like thinking from computers will result in
disappointment. However, before finalizing the data, a specialist review team typically examines
a large amount of it, and still, there are considerable differences in accuracy and consistency.
The subject matter expert gold standard, not the standard of the typical professional review team,
is expected to be confirmed by the accuracy required by the automated approach to document
classification. Professional review teams, rather than the subject matter experts, are coding most
of the information on legal document reviews. Thus, holding automated systems to a high
standard that is rarely, if ever, achieved in actual circumstances generates an unreasonable and
unattainable aim.
Legal AI is the application of technologies such as natural language processing, machine
learning, recognition of speech, legal robotics, preparation, natural image comprehension, rules-
based expert systems, neural networks for logical programming, artificial vision that was
machine learning, and neural networks to legal issues. Artificial intelligence (AI) has appeared
as one of the innovative methods to acquire information and knowledge in several sectors. AI
ANVESAK
ISSN : 0378 – 4568 UGC Care Group 1 Journal
Vol. 53, No.2 July – December 2023 58
can Flawlessly support the need to automate business processes, gain insight from data analysis,
and engage with customers and workers.
EVOLUTION
When Alan Turing began to show that the machine could think and learn by itself, thinking like
human beings became a central issue for scientists. After a series of tests on whether "machines
can think," Alan Turing could implement his hypothesis. Teaching machines to reflect and learn
like humans is now not impossible; various tests have revealed the same. Then, in 1956, at the
first academic conference on artificial intelligence, John McCarthy proposed the idea of artificial
intelligence, which outlined that artificial intelligence is about letting a machine simulate the
intelligent behavior of humans as precisely as it can.
The European Union adopted AI technology in 2001. EU has developed Legislation
Interoperability Tools LegIT, open software for drafting and editing legislation. Also, Article
36 of Turkey's Constitution requires adopting IT technology in the justice system. As a result,
the Government of Turkey has developed an advanced system of AI through the National
Judiciary Informatics System (UYAP). The Ministry of Justice (MOJ) IT Department developed
UYAP using the most updated technology and methodologies (EU, 2009). Currently, many
nations across the world have been modernizing the justice system by using IT under the
Strategic Plan of Judiciary
INTERPLAY OF ARTIFICIAL INTELLIGENCE IN LAW
Artificial intelligence (AI) has significantly impacted the legal sectors in the twenty-first century.
AI technology has assisted various legal professionals, including lawyers, judges, legislators,
and others, in securely storing many private files. The introduction of CD-ROM, electronic files,
and digital libraries drew lawyers towards digitalization to minimize human labor in text
searching and time difficulty in dealing with analog items. Lawyers and clients began
researching Google, digital libraries such as Lexis Nexis, Bloomberg, Justia, and CanLII
(Canada), and government documents such as court regulations, laws, and judicial decisions.
On the other hand, it saves time, assures efficiency, ensures accuracy, aids in document review,
conducts legal research, and so on. AI has recently become an essential topic for government
agencies, including the legislature. The laws have been posted on the websites of different
government agencies and made available to individuals worldwide. Similarly, law companies
worldwide are now employing AI to help their customers with counseling, document analysis,
and even filing court applications. The significance of AI in law is growing by the day, as seen
by the construction of National Judiciary Informatics Systems (UYAP) in numerous nations. As
a result, AI has played an essential role in the legal field. Despite this, it may do more damage
to the general public owing to various factors such as malware, data loss, and more. As a result,
to reap the benefits of AI in the subject of law, it must be employed with caution.
Analog resources such as educational materials, casebooks, case reporters, quickly loose-leaf
offerings, law journals, and laws were part of every lawyer's library during the evolution stage
of law. Because everything was in paper form, managing these records was tiresome. Their
storing, conserving, consulting, and sorting took a long time.
ANVESAK
ISSN : 0378 – 4568 UGC Care Group 1 Journal
Vol. 53, No.2 July – December 2023 59
The latest interest of lawyers is in computational technologies such as predictive tools and natural
language processing, which allow users to extract and reclaim relevant text from noisy data using
engines for searching, speech-to-speech translations, and intelligence aids for consumer
reimbursement. Artificial intelligence, in particular, is expected to disrupt and take over
consumer comprehension, implementation of company infrastructure management, and legal
processes. AI can do analytical activities and replicate most of the work now performed by
humans. The ability to incarcerate and manage inferred legal expertise that aids in decision-
making by lawyers and those they represent is critical here. Such systems function as a legal
storehouse, accumulating and gaining insight from their knowledge to better the advice provided.
Despite artificial intelligence's multiple problems, legal AI challenges human knowledge by
providing legal services through legal data research, prediction technology, e-discovery,
intelligent interfaces, triage services, and legal bots. Legal data research includes a prediction
system in which legal artificial intelligence predicts the outcome of an investigation based on a
specific theme, as well as/or lawsuit leaning depending on the outcome and knowledge system
of authorized research along practice lines.
TOOLS OF ARTIFICIAL INTELLIGENCE USED IN LAW
Legal AI comprises intelligent interfaces to assist lawyers in completing legal activities, contract
analysis to assist individuals in finalizing contracts by thoroughly analyzing them, and legal data
research to analyze legal data. Legal AI employs machine learning to accomplish these duties,
which entails inputting vast information, learning information by machine during the training
phase, and producing a learning-based output.
Litigators use planning and predictive Systems in pre-litigation preparations. This AI-powered
software examines many publicly available court qualifications, cases, and verdicts completed
by the arbiter in precedent up to the present day related to the case, as well as an innovative range
of valuable community statistics. A predictive system analyses data such as damage/costs
granted, cases resolved by specific firms, success/failure charges of certain lawyers, petition
winner/collapse, appealing/losing point of view, and judicial views/rulings. Its primary goal is
to reduce the size of a labor-intensive study and provide attorneys and clients with an enforceable
approach to previous cases. The deed of attorneys on similar matters, and, where possible, confer
some clue of reimbursement that could be granted by such affair/and/or additional feel value data
concerning the root of its gathered evidence regarding the liable triumph of an issue compared
to before similar matters.
Further, there is Lex Machina, which offers timing analytics using artificial intelligence to
forecast the estimated time of a case trial before a given judge. The most basic technique for
analysis is artificial intelligence, which discovers the best form from training data, which can
include legal documents, photos, audio, and more. Regarding legal AI, we have a legal expert
system, which is an analytical form based on legal expertise and competency, thus operating as
an e-lawyer. Division of Machine learning classification is into two types: supervised learning,
which teaches a representation of recognized input/output data to predict future outcomes amid
ambiguity, and unsupervised learning, which discovers hidden prototypes or basic information
in input data.
ANVESAK
ISSN : 0378 – 4568 UGC Care Group 1 Journal
Vol. 53, No.2 July – December 2023 60
The trial-and-error method selects the best algorithm for regression and classification in legal
situations. The method is because highly elastic algorithms tend to overfit data by introducing
noise. Choosing the best algorithm necessitates sacrificing one advantage in favor of another,
considering complexity, accuracy, and speed. MATLAB provides tools to let the person conduct
a variety of appliance learning algorithms and achieve the best results.
SVM, Discriminant Analysis, nave Bayes, closest neighbor, and neural networks are some
techniques available in MATLAB for classification. For ensemble methods, decision trees and
neural networks perform well for regression techniques such as linear regression, GLM, SVR,
and GPR. MATLAB also has regression and classification learner apps that help us choose the
optimal model. These programs analyze the data, choose features, and visualize the results,
preventing overfitting via holding out or cross-validation. Using these tools and strategies, we
may organize our data in an organized manner by categorizing it based on litigations, decisions,
judgments, and appeals on cases.
CONTRACTUAL HELP: To assist people in finalizing contracts, legal AI first scans and then
analyses legal agreements such as leases and commercial contracts to extract meaningful data
and compare them to current laws/rules. The fuzzy rule-based system is the most essential soft
computing technique for decision-making. As the base of this AI system is Natural Language,
the Fuzzy approach performs well when dealing with imprecise data. It uses fuzzy if-then rules
to transform the human answer into fuzzy dependence and a common language. Leasing law,
eDiscovery, diligence checks, sales/procurement contract review, risk and compliance
assessment, financing/ OTC derivative deal analysis, and employment contract review are some
of the uses discovered by law firms.
Due Diligence
Legal Robot, JPMorgan, Judicata: Litigators employ AI
techniques to undertake due diligence to identify
background information. It aids in counseling customers
on the available options in a legal scenario and the
essential action to resolve the issue.
Prediction Technology
To find the anticipated outcome of litigations, litigators
use artificial intelligence algorithms such as Everlaw,
DISCO, Catalyst, Exterrro, Brain space discovery,
Intraspexion, and Premonition.
Legal Analytics
Lawyers can use information based on previous case
win/loss history, previous case law, and judge's history to
identify patterns and trends. Lex Machina, Ravel Law
Document Automation
The report, perfect NDA, Law firms employ software
templates to construct filled-out papers based on data
inputs.
ANVESAK
ISSN : 0378 – 4568 UGC Care Group 1 Journal
Vol. 53, No.2 July – December 2023 61
Intellectual Property
Trademark Now, ANAQUA Studio, Smart Shell, and
Lawyers employ AI tools to help them analyze extensive
IP portfolios and make conclusions from the context.
Electronic Billing
Legal AI assists lawyers in automatically computing
billable hours: Bright Flag and Smoke Ball.
A SCOPE FOR IMPROVEMENT
Several steps can be put in place to lessen the abuse of AI in the Indian legal system:
1. Robust Regulations: India just implemented regulations for the availability and misuse of
digital data via the Digital Data Protection Bill 2023, and the country already has IT regulations.
Implementation of such regulations simply means that India is already halfway there. Now, all
the country needs is detailed rules governing AI's application to the legal system. This legislation
should cover algorithm accountability, openness, and the legal limits of AI's influence on
decision-making. Adopt strict data protection controls to shield sensitive legal data against
manipulation, breach, or unauthorized access.
2. Transparency Reports: Establishments utilizing AI in the court system should be required
to regularly produce transparency reports outlining the technology's application, effectiveness,
and any remedial measures implemented. Such transparency encourages responsibility and
public confidence.AI algorithms employed in legal proceedings must also be released in public,
as done by Poland in 2021, after some judges complained about the error in the AI system used
by the Ministry of Justice. To reduce biases and errors, ensure the algorithms are thoroughly
tested, audited, and verified. Accountability mechanisms are required to hold developers
accountable for any unfavorable effects.
3. Ethics Committees: Establishment of impartial Quasi-Judicial ethics committees with legal
and AI experts and ethicists to ensure that AI technologies in the judicial system uphold moral
and legal standards. These committees can examine and authorize the Usage of Artificial
Intelligence. The Committee will also regularly evaluate and modify AI policies and guidelines
to reflect new technological developments and evolving ethical concerns.
4. Continuous Monitoring and Auditing: While using any AI, Consistently examining and
auditing is needed to find and fix any biases, mistakes, or unintended outcomes that may occur.
Such monitoring will result in preserving the honesty and equity of legal proceedings.
5. Human Oversight: No matter how accurate and Effective AI can be, it will always need more
empathy and critical thinking. In 2020, some renters were wrongly listed as sexual Offenders,
eventually blocking them from getting a home. These crises can be averted if there is a slight
human insight into these AI-generated suggestions or decisions. Continue to exercise a sizable
amount of human judgment in critical legal issues involving AI.
6. Education and instruction: Give legal professionals specialized instruction on artificial
intelligence (AI), its potential, ethical considerations, and limitations. They can use AI
ANVESAK
ISSN : 0378 – 4568 UGC Care Group 1 Journal
Vol. 53, No.2 July – December 2023 62
technologies and analyze their outcomes with more knowledge. Start public education efforts to
inform people about the legal system's use of AI and its potential limitations. Training will
remove any misconceptions about AI and help demystify it.
7. User Consent: Verify that parties to judicial proceedings are aware of the use of AI and how
it can affect their cases. Before using AI in decision-making processes, get explicit consent.
There is also a dire need to extend the already given concept of consent in the new Data
Protection Billto make it more inclusive to all vulnerable stakeholders.
By putting these policies in place, India can take advantage of AI's advantages in its judicial
system while lowering the chance of abuse and supporting the ideals of fairness, equity,
and transparency.
CONCLUSION
Artificial intelligence strategies are gaining popularity due to increased client demands on
dominating businesses to be more proficient and a growing aversion to remuneration for what
they view to be advancement-stage toil. This paper covers current trends in legal AI and suggests
strategies that can be used in the future. In the future, using MATLAB software, one can perform
deep learning and machine learning algorithms on legal documents. We can create a tax
commandment classifier using classification and deep learning based on print published in the
Supreme Court, High Court, and district court verdicts in which the courtyard announced a
twofold resolution for the specific legal matter.
There are also many complex and critical ethical issues regarding the employment of AI in India's
legal system. Addressing issues about transparency, accountability, bias reduction, and job
displacement is critical as AI technologies advance and become more integrated into legal
procedures. Although artificial intelligence (AI) has an opportunity to improve effectiveness as
well as accessibility to justice, it needs to be deployed cautiously and under a strict legal
framework to prevent unforeseen consequences.
Furthermore, several recommendations might be taken into account to manage these ethical
problems successfully, and the author has discussed a few of them. However, explicit norms and
standards should be developed for AI algorithms employed in legal decision-making to ensure
openness and justice. Second, it should be required that AI systems be continuously monitored
and audited in order to spot and correct any biases that may develop over time. Third, specialized
training should be provided to legal professionals so they can comprehend AI's potential,
constraints, and ethical ramifications.
Furthermore, a balanced strategy reconciling technology breakthroughs with legal ideals requires
cooperation between legal professionals, AI developers, and policymakers. Public awareness
initiatives can reduce fears and encourage trust by informing people about AI's role in the legal
system. As AI develops and becomes more prominent in India's legal system, continued study
and discussion should remain a top focus.
REFERENCES
ANVESAK
ISSN : 0378 – 4568 UGC Care Group 1 Journal
Vol. 53, No.2 July – December 2023 63
[1] Huawei Technologies Co., Ltd. (2023). A General Introduction to Artificial Intelligence.
In: Artificial Intelligence Technology. Springer, Singapore. Retrieved from
https://doi.org/10.1007/978-981-19-2879-6_1
[2] Anyoha, R. (2017).‘The History of Artificial Intelligence in India’
https://sitn.hms.harvard.edu/flash/2017/history-artificial-
intelligence/#:~:text=One%20such%20person%20was%20Alan,machines%20do%20th
e%20same%20thing%3F
[3] Khan, A. (2023).AI-powered Indian judiciary: A step forward or cause for concern?
Retrieved from https://www.barandbench.com/columns/litigation-columns/ai-powered-
indian-judiciary-a-step-forward-cause-
concern#:~:text=During%20a%20hearing%20in%20February,Justice%20of%20India%
20DY%20Chandrachud
[4] Tate, K. (2014).‘History of AI: Artificial Intelligence’
https://www.livescience.com/47544-history-of-a-i-artificial-intelligence-
infographic.html
[5] Access to government algorithms in Poland Retrieved from
https://digitalfreedomfund.org/access-to-government-algorithms-in-poland/
[6] Kirchner, Lauren & Goldstein, M. (2020).How Automated Background Checks Freeze
Out Renters. Retrieved from https://www.nytimes.com/2020/05/28/business/renters-
background-checks.html
[7] Sapkota, T.P., Kunwar, S., Bhattarai, M., &Poude, S. (2020). Artificial Intelligence That
Are Beneficial for Law. US-China Law Review, 17 (5), 217-223
[8] Dabass, J., &Dabass, B.S., (2018). Scope of Artificial Intelligence in Law. Retrieved
from Preprints.org
[9] Hall, A. C. (2018). How law firms can benefit from artificial intelligence. Retrieved from
https://www.lawtechnologytoday.org/2018/11/how-law-firms-can-benefit-from-
artificial-intelligence/
[10] LexisNexis. (2018). New technology equips the legal field for the future. Retrieved from
https://www.golegal.co.za/legal-technology-future/
[11] Davenport, T. H., &Ronanki, R. (2018). Artificial intelligence for the real
world. Harvard Business Review. Retrieved from https://hbr.org/2018/01/artificial-
intelligence-for-the-real-world
[12] Marchant, G. E. (2017). Artificial intelligence and the future of legal practice. Retrieved
from https://www.iadclaw.org/assets/1/7/10.4-_Marchant-
_ai_and_practice_of_law_SciTech_lawyer.pdf