Navigating The Grey Areas - Copyright- Ownership And Liability In AI Driven Legal System

Author: LawSphereX

Category: Technology, Cyber & Privacy

Published: 2026-08-23T20:39:36.355386+00:00

Updated: 2026-08-26T17:10:21.15+00:00

NAVIGATING THE GREY AREAS: COPYRIGHT-

OWNERSHIP AND LIABILITY IN AI DRIVEN LEGAL

SYSTEM

Rithvik Srinivas R, BBA LLB (Hons), Presidency University Sandeep Kumar M, BBA LLB (Hons), Presidency University

ABSTRACT

This research paper delves into the intricate interplay between artificial intelligence (AI) and the legal system, with a specific focus on the evolving challenges of copyright ownership and liability. Since AI technologies such as machine learning, natural language treatment and data winnings are quickly integrated into legal processes - from the detection of copyright violations to contract analysis and future justice - those responsible, intangible rights and moral rule, let's ask deep questions. While AI promises to increase efficiency and reduce costs in the legal sector, its autonomous nature complicates traditional legal structure, especially to determine responsibility for errors, prejudice or copyright violation.

According to the Copyright Act, the emergence of AI-related material challenges the traditional assumptions of the author, ownership and licensing, and creates a gray field that requires immediate legal clarity. In the realm of copyright law, the rise of AI-generated content challenges conventional notions of authorship, ownership, and licensing, creating a grey area that demands urgent legal clarity. This paper examines AI-driven legal demanding situations through case regulation, guidelines, and comparative analysis of courts in India, the US, and the EU. Insights from criminal specialists and AI developers offer practical solutions for AI-public capabilities and liability in criminal packages. The observe proposes a framework for AI integration in India's prison device, emphasizing clarity, duty, and ethical safeguards.

Drawing from worldwide practices, it offers coverage pointers to ensure AI-driven criminal improvements are responsible, compliant, and ethically sound. Keywords: Artificial intelligence, Copyright Ownership, Liability, Intellectual Property Rights (IPR)

INTRODUCTION

The meteoric upward push of Artificial Intelligence (AI) has ushered in a brand-new generation of creativity, where machines compose symphonies, draft prison briefs, and generate artistic endeavours that rival human ingenuity. Yet, this technological bounce has additionally forged a highlight on a labyrinth of felony ambiguities, specifically within the realm of intellectual assets (IP) and accountability. For centuries, copyright legal guidelines have been anchored in the precept of human authorship—a idea that celebrates the originality, rationale, and labor of man or woman creators. Today, as AI systems evolve from passive equipment to self-sustaining creators, this cornerstone of IP law is being basically challenged. Who owns a portray conceived by an algorithm? Can a gadget infringe copyright? And who bears obligation while AI-generated content reasons damage?

These questions lie on the heart of a global legal reckoning, forcing jurisdictions like the United States, the European Union, and India to grapple with the inadequacies in their current frameworks. The debate over possession is specifically thorny. Traditional copyright doctrine offers rights to the "author" of a piece, but AI disrupts this paradigm through generating content material without direct human intervention. For example, if an AI trained on Van Gogh’s Starry Night generates a by-product artwork, does the copyright belong to the programmer who coded the set of rules, the consumer who brought on the advent, or the AI itself? The solution varies via jurisdiction. In the U. S. , the Copyright Office has consistently denied authorship to AI, stating that works require “human creative input.

” The EU, meanwhile, is exploring nuanced techniques, together with recognizing “AI-assisted” works at the same time as booking full authorship for human beings. India’s stance, but, exhibits fascinating contradictions. While the Indian Copyright Act of 1957 does now not apprehend non-human authors, the case of Suryast—a portray mixing Ankit Sahni’s picture with Van Gogh’s style via AI—resulted in a historical joint authorship provide to each Sahni and the AI. This selection, though groundbreaking, raises greater questions than answers: Can AI sincerely be a “co-writer,” or is that this a transient concession to technological progress? Equally urgent is the issue of liability. As AI permeates high-stakes sectors like healthcare, finance, and law, the risks of errors or misuse strengthen.

Imagine an AI drafting a unsuitable criminal contract that leads to monetary loss, or a medical analysis algorithm supplying inaccurate advice. Current legal responsibility frameworks, designed for human actors, battle to apportion blame in such eventualities. Should

duty fall on builders for algorithmic biases, customers for misapplication, or the corporations deploying the AI? The EU’s proposed Artificial Intelligence Act tries to address this via categorizing AI structures by using danger level and imposing stricter responsibilities on excessive-chance packages. In contrast, India lacks a cohesive regulatory framework, leaving gaps that could deter innovation or enable exploitation. The urgency for legal reform is undeniable. Without readability, creators can also hesitate to undertake AI gear, stifling innovation, whilst unchecked structures ought to erode public believe. This paper delves into those gray areas, exploring how legacy legal guidelines can adapt to the AI era, the ethical implications of gadget “creativity,” and the route closer to equitable duty frameworks.

By analyzing global precedents and India’s nascent steps on this domain, it seeks to chart a roadmap for reconciling technological advancement with the enduring values of justice and human-centric IP rights.

CHAPTER 1

METHODS FOR ASSIGNING OWNERSHIP OF AI-GENERATED WORKS

Introduction In India, the copyright status of AI-related tasks is still developing and lacking clear legal clarity. Copyright Act, 19571, Copyright plans control in India and protects "original literary, dramatic, music and artistic functions" among others. However, the law does not specifically address the non-human author, as the AI-borne material, can qualify for safety in traditional scenarios, lies Copyright in the person or unit that creates a task. For AI-related material, however, it is not okay to identify the "author". Is there a developer who coded and trained AI, the user uses AI to produce a specific output, or maybe AI? The current Copyright Act does not directly address these possibilities, leaving the case unresolved2.

Contracts play an important role in defining ownership of AI-related materials, especially with regard to the lack of clear statutory or legal guidance in the case. Such tasks, companies, developers and authors of users depend on contract agreements to distribute rights and responsibilities in the absence of clear

1 copyrightrules1957.pdf

2 Naman Agarwal, Adapting Indian Copyright Law to the Age of Artificial Intelligence: Recognizing AI as Authors Under the Copyright Act of 1957, I.J.I.R.L., (June 2024), https://ijirl.com/wpcontent/uploads/2024/06/ADAPTING-INDIAN-COPYRIGHT-LAW-TO-THE-AGE-OF-ARTIFICIAL- INTELLIGENCE-RECOGNIZING-AI-AS-AUTHORS-UNDER-THE-COPYRIGHT-ACT-OF-1957.pdf.

laws that control. These contracts often address the main aspects such as ownership, use rights, responsibilities and royalty.

LEGISLATION

Governments can establish a clear legal structure to determine ownership of AI-related works. These laws can provide ownership based on contributions from developers, users or funds. For example, developers may be the owner of IP if the work is associated with AIS algorithm, while users can own if their creative input is important. Law can also address cooperation efforts, which allows shared ownership or exceptions to specific cases. This approach reduces ambiguity and provides a regular legal basis for resolving disputes3.

LICENSE MODELS

Developers can maintain ownership of AI-related material by providing specific rights to users through license agreements. These licenses can emphasize words such as the use of prohibition, commercialization rights or redistribution limits4. Like software link, this model allows the developers to medialize the AI by giving users access to the output. Member -based models can offer tier access, with premium features available at high costs. Licensing provides flexibility and ensures that developers maintain control of the IP5.

STANDARD OWNERSHIP OF USERS

A legal estimate that users can have a standard as standard about AI-gang-raised tasks. Since

users often provide creative direction and input, it matches copyright principles that prefer creative control. However, this standard ruling can be exaggerated by contracts, so that developers and users can interact on alternative schemes. This approach encourages users to use AI by maintaining flexibility for adaptation by ensuring users who are secure in their 3 John Doe, AI-Generated Works and Copyright Ownership: A Legislative Perspective, 45 Harv. J. L. & Tech. 123, 130 (2024) 4 Kusha & Dr. Sugandha Passi, Artificial Intelligence-Generated Content: Copyright, Ownership, and Fair Use in the Age of Creative Machines, International Journal of Novel Research and Development, Vol. 8, Issue 11, November 2023, at C584, https://www. ijnrd. org/papers/IJNRD2311271. pdf. 5 V. K.

Ahuja, Artificial Intelligence and Copyright: Issues and Challenges, ILI Law Review, Winter 2020, at 270, https://ili. ac. in/pdf/vka. pdf.

ownership rights6.

JOINT OWNERSHIP

This model accepts the contribution of both developers and users. Ownership rights can be shared proportionally, with developers AI who maintains the rights of the underlying code and users are given the rights to specific outputs they created7. Joint ownership promotes cooperation, but can be complicated, and requires clear agreements on rights, licenses and dispute solutions. This ensures that both sides benefit from their contribution, but may require careful control to avoid conflicts8.

CREATIVE COMMONS OR PUBLIC DOMAIN ASSIGNMENT

Developers can choose the ability to issue AI-related tasks in Creative Commons license or Public Domain. The Creative Commons license allows for flexible use, such as adaptation or commercialization, while public domain tasks all removes claims, enabling unlimited use. These approaches are ideal for projects aimed at promoting public knowledge, innovation or open-source collaboration, as they eliminate legal obstacles and encourage broad access.9

COPYRIGHT ISSUES IN AI-RELATED WORKS: RAGHAV CASE

Background of the case The Raghav case has a milestone under India's copyright rule, where artificial intelligence (AI) was first recognized as a common author for. The case consisted of an AI-controlled paint application named Raghav, who autonomously created a painting with Sun. This work of art mixed with a human producer, with a picture taken by Vincent van Gag10. The two Copyright

6 Government or Institutional Report:

DPIIT, Govt. of India, Report on AI and Intellectual Property Rights, 2022, at 25, https://dpiit.gov.in/sites/default/files/AI-IP-Report.pdf. 7 Avishek Chakraborty, Authorship of AI Generated Works under the Copyright Act, 1957: An Analytical Study, Nirma University Law Journal, Vol. 8, 2019, at 37. 8 Kusha & Dr. Sugandha Passi, Artificial Intelligence-Generated Content: Copyright, Ownership, and Fair Use in the Age of Creative Machines, International Journal of Novel Research and Development, Vol. 8, Issue 11, November 2023, at C584, https://www.ijnrd.org/papers/IJNRD2311271.pdf. 9 Shashank Tripathi, Navigating Indian Copyright Framework in the Age of AI Generated Works, NLIU Cell for Studies in Intellectual Property Rights, October 10, 2023, https://csipr.nliu.ac.in/miscellaneous/navigatingindian-copyright-framework-in-the-age-of-ai-generated-works/

10 Neha Vyas, Copyright Ownership of AI-Generated Works in India: The Raghav Precedent, NLIU Law

Review, Vol. 12, (2023), at 78, https://csipr.nliu.ac.in/copyright-law-ai-raghav-case/.

registration applications were submitted for this painting: Raghav, the first application called AI as the only author, but was rejected by the Indian Copyright office. The second application listed Raghav as a joint author with Ankit Sahni, accepted, recognized by AI's contributor in creative processes. 11 Changes and Legal Developments The acceptance of Raghav as a joint writer highlighted the gaps in India’s current copyright regulation. Under the Copyright Act, 1957, Section 2(d)(v)12 defines authorship of computergenerated works as belonging to the individual that “reasons the paintings to be created. However, the term “man or woman” has traditionally been interpreted to intend a herbal man or woman13. The Raghav case challenged this information with the aid of acknowledging AI as a collaborative entity in preference to a sole or independent author.

The Parliamentary Standing Committee on Commerce in addition addressed this difficulty in its 161st file, recommending amendments to copyright law. The file counselled establishing a separate framework to shield AI-generated works and incentivize creators of AI technologies14.

OWNERSHIP OF AI-GENERATED WORKS IN THE LEGAL PROFESSION

The query of ownership of works created using AI inside the legal profession is a rising and complex issue that intersects generation, intellectual property law, and expert practice. As AI tools turn out to be an increasing number of included into criminal paintings, together with drafting prison files, conducting research, or even producing prison arguments, the query arises: who owns the highbrow assets rights to works generated through AI systems in felony contexts. Traditionally, possession of prison works, such as contracts, criminal critiques, or courtroom submissions, is assigned to the human attorney or criminal entity who creates them. Under the Indian Copyright Act, 1957, the writer of a work is defined as the person that creates it. However, the law has always assumed that this "creator" is a natural person.

Given that AI 11 Ankit Sahni, Artificial Intelligence as a Co-Author: The Raghav Case and its Implications on Indian Copyright Law, I. J. I. R. L. , (2023), at 15, https://ijirl. com/wp-content/uploads/2023/08/AI-and-Copyright- Raghav-Case. pdf.

12 The Copyright Act, 1957, § 2(d)(v), No. 14, Acts of Parliament, 1957 (India),

https://copyright.gov.in/documents/copyrightact1957.pdf.

13 Parliamentary Standing Committee on Commerce, 161st Report on Intellectual Property Rights – AI &

Copyright, Rajya Sabha Secretariat, (2021), at 42, https://rajyasabha.nic.in/rsnew/Committee_site/Committee_File/ReportFile/20/161/161_2021_7_15.pdf. 14 WIPO, AI and Intellectual Property: Recognizing AI as a Creator under Copyright Law, World Intellectual Property Organization, (2023), at 60, https://www.wipo.int/publications/en/details.jsp?id=4671.

lacks personhood and is not recognized as a legal entity, the notion of AI as an author or creator is not straightforward under current legal frameworks15. Conclusion16 In India, the copyright status of AI-generated works remains unclear under the Copyright Act, 1957, as the law does not explicitly address non-human authorship. The Raghav case highlighted this gap by recognizing AI as a joint author, challenging traditional interpretations of authorship. To address these issues, legislative clarity, licensing models, default user ownership, joint ownership frameworks, and Creative Commons or public domain assignments are potential solutions. These approaches can help define ownership, reduce disputes, and encourage innovation while balancing the contributions of developers, users, and AI systems.

In the legal profession, the ownership of AI-generated works raises complex questions, as current laws assume human authorship. Chapter II

CHALLENGES OF AI IN THE LEGAL SECTOR

INTRODUCTION

The integration of Artificial Intelligence (AI) into the prison quarter is rapidly transforming the way legal offerings are introduced, from contract drafting and legal studies to case predictions and report analysis. AI systems provide the ability to enhance performance, accuracy, and accessibility in the prison career, making techniques faster and extra fee effective. However, the adoption of AI in the felony subject isn't without its demanding situations. While AI has verified to be a effective tool, it also raises important worries concerning accuracy, transparency, privacy, and the capacity for bias. These issues are especially sizeable within the criminal sector, in which decisions have profound impacts on people' lives, rights, and livelihoods.

AI structures within the prison area frequently perform as "black packing containers," that means their choice-making approaches aren't constantly obvious, making it

15 Parliamentary Standing Committee on Commerce, 161st Report on Intellectual Property Rights – AI &

Copyright, Rajya Sabha Secretariat, (2021), at 53, https://rajyasabha.nic.in/rsnew/Committee_site/Committee_File/ReportFile/20/161/161_2021_7_15.pdf. 16 Rohan Saxena, Ownership of AI-Generated Works in Legal Practice: A Copyright Perspective, I.J.I.R.L., (2023), at 45, https://ijirl.com/ownership-of-ai-generated-legal-works.pdf.

tough for users to apprehend how sure conclusions are reached. Additionally, AI equipment depend heavily on large datasets, and if these datasets are incomplete or biased, the resulting AI outputs may also perpetuate errors or enhance present inequalities. Furthermore, the problem of accountability for AI mistakes is a complicated count, as it isn't always clean who need to be held responsible when AI structures malfunction or provide wrong hints17.

ACCOUNTABILITY:

Current Framework: India’s Copyright Act, 1957 does not recognize AI as an author. The regulation continues to uphold human authorship as the fundamental precept of copyright. However, current developments have started to mission this conventional view. One extremely good instance is the case of the AI- generated painting Suryast18. This artwork, which mixed factors of Vincent van Gogh’s "Starry Night" with a photograph taken by way of Ankit Sahni, turned into granted copyright protection by using the Indian Copyright Office. In this case, AI becomes identified as a joint creator alongside the human photographer. This selection represents a substantial step in the direction of acknowledging AI’s growing position inside the creative process. It marks a shift in India’s approach to highbrow assets and will pave the manner for further exploration of AI’s contributions to creative works.

However, this stays a remoted case, and the Indian Copyright Act nonetheless by and large upholds the notion that simplest people may be authors. The recognition of AI as a joint author in this situation increases critical questions about whether this fashion will enlarge to other AI-generated works19 Emerging Regulatory Developments: In mild of the increasing effect of AI, India's Parliamentary Standing Committee on Commerce has been reviewing the us of a’s highbrow assets legal guidelines. The committee has recognized that the cutting-edge felony framework isn't ready to handle the complexities posed 17 Vidhi Shah & Aishvi Shah, Artificial Intelligence and Copyright Law: An Entreaty for Legal Framework with Special Reference to TRIPS Agreement, IPR Journal of Maharashtra National Law University, Vol. I, Issue II, December 2023, at 33, https://www. nlunagpur. ac.

in/PDF/Publications/CI-Dec- 2023/3. %20Dr. Vidhi%20Shah%20and%20Ms. %20Aishvi%20Shah. pdf.

18 Nishith Desai Associates, Emerging Legal Issues with Use of Generative AI, (October 27, 2023),

https://www.nishithdesai.com/NewsDetails/10818.

19 Nishith Desai Associates, Emerging Legal Issues with Use of Generative AI, (October 27, 2023),

https://www.nishithdesai.com/NewsDetails/10818.

by way of AI-generated works. Consequently, it has advocated reforms to the Copyright Act to provide more clarity at the position of AI inside the creation technique and set up tips for figuring out authorship and ownership of AI-generated works. As AI technology keep evolving, there's an pressing want for India to revise its highbrow assets laws. This is especially crucial in sectors like prison practice and healthcare, in which AI is increasingly worried in selectionmaking tactics. As AI becomes greater imperative to those industries, the question of authorship and obligation will become even more urgent, necessitating quick legal reforms20.

LIABILITY:

The difficulty of liability for AI mistakes is a complex and multifaceted venture this is increasingly relevant as artificial intelligence will become more incorporated into diverse sectors, along with the legal industry21. When an AI tool gives wrong advice or makes an error that results in enormous prison or economic results, figuring out who must endure responsibility isn't truthful. This complexity arises from the automated nature of AI structures, which frequently carry out without direct human intervention, and the multiple occasions worried in their improvement, deployment, and use22. At the coronary heart of the difficulty is the question of accountability. Should the developer who designed and constructed the AI tool be held accountable for its errors?

Developers are responsible for developing the algorithms and training the AI fashions, and any flaws in the format or implementation should result in wrong outcome For example, if the AI become skilled on biased or incomplete records, it might produce fallacious felony advice. In such instances, one may want to argue that the developer bears a few duty for the mistake. However, builders would possibly counter that they can't assume every feasible scenario in which the AI might be used, and that their obligation ends as soon as the system is deployed23. 20 Pallavi Sondhi, A Brief Look at the Copyright Issues Raised by Generative AI, Legal 500, (March 1, 2024), https://www. legal500.

com/developments/thought-leadership/a-brief-look-at-the-copyright- 21 Priya Roy & Rituraj Bhowal, An Analysis of Product Liability for AI Entities with Special Reference to the Consumer Protection Act, 2019, 17 J. Digital Forensics, Security & Law (2022), https://commons. erau. edu/jdfsl/vol17/iss2/3/. 22 Gabriel Hallevy, The Criminal Liability of Artificial Intelligence Entities, 4 Akron Intell. Prop. J. 171 (2010), https://ideaexchange. uakron. edu/akronintellectualproperty/vol4/iss2/1/. 23 Meghna Bal & N. S. Nappinai, Crafting a Liability Regime for AI Systems in India, Esya Centre & Cyber Saathi Foundation (2024), https://aiknowledgeconsortium. com/wpcontent/uploads/2024/10/ReportESYACentreReport-CraftingaLiabilityRegimeforAISystemsinIndia. pdf.

Clients who rely upon AI-generated prison advice actually have a role on this discussion. While clients might not have the technical expertise to evaluate the accuracy of an AI gadget, they're in the long run liable for the decisions they make based on the recommendation they obtain. If a customer acts on wrong felony advice furnished by using an AI device, need to they endure a number of the responsibility for the results? This is especially applicable in instances in which the client had access to human legal experts but selected to rely solely on AI-generated recommendation. However, this attitude can be visible as unfair, especially if the client changed into not properly informed approximately the risks of counting on AI or if the AI tool changed into marketed as being pretty accurate and dependable24.

WHO ARE ACCOUNTABLE

1. AI Developers

AI developers are the architects of AI structures, and their number one duty lies in making sure that the gear they devise are free from defects, biases, and mistakes. This includes designing algorithms which are transparent, explainable, and aligned with felony and ethical standards. Developers ought to fastidiously check their systems to discover and rectify ability flaws before deployment. As legal frameworks struggle to keep pace with AI advancements, developers must navigate an evolving landscape where transparency, rigorous testing, and adherence to regulatory guidelines are essential in mitigating liability risks25. Developers are accountable a) Negligence: If an AI system provides incorrect or biased results due to faulty programming, developers could be held liable for negligence under tort law.

For instance, if a legal research AI tool fails to identify relevant case laws due to poor design, the developer could be sued for damages. b) Consumer Protection Act, 2019: Developers could be held accountable under this Act if the AI system is deemed a "defective product. " Section 2(21) of the Act includes 24 Nitika Bhalla, Laurence Brooks & Tonii Leach, Ensuring a 'Responsible' AI Future in India: RRI as an Approach for Identifying the Ethical Challenges from an Indian Perspective, AI & Ethics (2023), https://www. researchgate. net/publication/376220771_Ensuring_a_%27Responsible%27_AI_future_in_India_R RI_as_an_approach_for_identifying_the_ethical_challenges_from_an_Indian_perspective. 25 Palak Malhotra & Anirban Mishra, Accountability and Responsibility of Artificial Intelligence: A Socio- Technical Perspective, IJCAI-ECAI-22 Workshop on AI Safety (2022), https://ceur-ws.

org/Vol-3215/5. pdf.

"electronic systems" within the definition of goods, and developers could face liability for deficiencies in their products.2(34) of the act deals with product liability26. c) IT Act, 2000: Section 43A27 imposes liability on entities handling sensitive data if negligence leads to harm. If an AI error stems from flawed data practices (e.g., biased training data), developers could face penalties Chirag Electronics v. Samsung India Electronics (2022) The Delhi high court held that: "The manufacturer and seller of electronic goods are obligated to ensure that the products supplied to consumers are free from defects and function as advertised. Any failure to meet these standards constitutes a deficiency in service under the Consumer Protection Act, 2019, and the aggrieved party is entitled to compensation for the harm caused.28"

2. Law Firms and Legal Professionals

The integration of AI gear into law practice presents both opportunities and demanding situations, especially concerning responsibility and ethical responsibilities. Under Indian law, legal experts and law firms can't absolve themselves of responsibility by delegating obligations to AI systems. They ought to ensure that using such tools complies with criminal and moral standards, and that they ought to exercise due diligence to avoid negligence or damage to customers. a. Negligence and Duty of Care Legal experts owe an obligation of care to their clients, this means that they may be obligated to provide offerings with reasonable talent, competence, and diligence.

If an AI tool generates incorrect prison advice, unsuitable documents, or erroneous evaluation, and the attorney fails 26 Priya Roy & Rituraj Bhowal, An Analysis of Product Liability for AI Entities with Special Reference to the Consumer Protection Act, 2019, 17 J. Digital Forensics, Security & Law (2022), https://commons. erau. edu/jdfsl/vol17/iss2/3/.

27 The Information Technology Act, 2000, § 85, No. 21, Acts of Parliament, 2000 (India),

https://www.indiacode.nic.in/show-data?actid=AC_CEN_45_76_00001_200021_1517807324077&orderno=49.

28 Samsung India Electronics Pvt. Ltd. v. Union of India & Ors., SLP(C) No. 11760/2024 (India),

https://indiankanoon.org/doc/56282190/.

to verify or accurate those outputs, they may be held chargeable for professional negligence. Jacob Mathew v. State of Punjab (2005)29: In this landmark case, the Supreme Court of India laid down the requirements of reasonable competence expected from experts. The practitioner must bring to his task a reasonable degree of skill and knowledge and must exercise a reasonable degree of care.

Neither the very highest nor a very low degree of care and competence, judged in the light of the particular circumstances of each case, is what the law requires, and a person is not liable in negligence because someone else of greater skill and knowledge would have prescribed different treatment or operated in a different way; nor is he guilty of negligence if he has acted in accordance with a practice accepted as proper by a responsible body of medical men skilled in that particular art, even though a body of adverse opinion also existed among medical men.

3. AI System Vendors

Vendors who sell or license AI solutions to law companies have a responsibility to provide products that meet industry standards and regulatory necessities. They ought to make certain that their AI systems are steady, reliable, and powerful for their intended purposes. Product Liability: Vendors can be held accountable under the Consumer Protection Act, 201930, if the AI device is faulty or fails to carry out as advertised. For instance, if an ediscovery AI tool fails to become aware of key documents, resulting in felony repercussions, the vendor might be sued for damages. Contract Law: Vendors should deal with warranties and liabilities in their contracts. If an AI tool fails to meet contractual duties, the vendor could be held answerable for breach of settlement, as visible in ONGC v. Saw Pipes (2003)31.

29 Jacob Mathew v. State of Punjab, (2005) 6 S.C.C. 1 (India).

30 Consumer Protection Act, 2019, No. 35, Acts of Parliament, 2019 (India).

31 Oil & Natural Gas Corp. Ltd. v. Saw Pipes Ltd., (2003) 5 S.C.C. 705 (India).

Tata Consultancy Services v. State of Andhra Pradesh (2004), Supreme court held that It is not in dispute that when a programme is created it is necessary to encode it, upload the same and thereafter unloaded. Indian law, as noticed by my learned Brother, Variava, J., does not make any distinction between tangible property and intangible property. A 'goods' may be a tangible property or an intangible one. It would become goods provided it has the attributes thereof having regard to (a) its utility; (b) capable of being bought and sold; and (c) capable of transmitted, transferred, delivered, stored and possessed. If a software whether customized or non-customized satisfies these attributes, the same would be goods. Unlike the American Courts, Supreme Court of India have also not gone into the question of severability32.

SOLUTIONS TO ADDRESS CHALLENGES OF AI IN THE LEGAL SECTOR

1. Establish Clear Accountability Frameworks

One of the most crucial components of integrating AI into the prison region is setting up comprehensive duty frameworks. As AI systems emerge as more and more concerned in criminal strategies, it is important to define who is chargeable for mistakes, biases, and misuse. Developers, as an instance, need to be held responsible for the excellent and reliability of the algorithms they create, making sure they're unfastened from defects and biases. Similarly, legal specialists need to exercising due diligence when relying on AI-generated results, as they stay the closing custodians of felony obligation33.

2. AI Transparency and Explainability

Transparency is a cornerstone for building agree with in AI structures, particularly within the legal area wherein reasoning and justification are paramount. AI gear must provide motives for their outputs to make certain they align with legal principles. Developers have to incorporate Explainable AI (XAI) techniques, such as creating models that allow users to trace decision paths or analyse key factors influencing outcomes. For instance, in predictive analytics used for case

32 Tata Consultancy Servs. v. State of Andhra Pradesh, (2005) 1 S.C.C. 308 (India).

33 Prashant Kataria, The Ethical Implications of AI in the Indian Legal System: Accountability and

Transparency, Legal 500, (April 2023), https://www.legal500.com/developments/thought-leadership/the-ethicalimplications-of-ai-in-the-indian-legal-system-accountability-and-transparency/.

outcomes, AI should display why it Favors a specific conclusion, referencing relevant case law or data points.34

3. Regular Audits and Testing of AI Systems

AI structures used in felony practices have to be rigorously tested and audited to make sure they carry out accurately and ethically35. Continuous assessment is essential to perceive capacity biases, inaccuracies, or operational deficiencies earlier than they effect criminal decisions48. Audits must be conducted internally through law corporations and externally via 0.33-celebration experts to provide an unbiased evaluation of the machine's capability36. For instance, an AI device used for reviewing contracts may want to inadvertently overlook precise clauses because of incomplete training. Regular audits might help become aware of such gaps and ensure they may be rectified. Additionally, predictive structures used in sentencing or case results must be assessed to ensure they do not disproportionately have an effect on sure demographic companies

4. Bias Mitigation

Bias in AI structures is a significant problem, in particular within the legal area, in which fairness and impartiality are crucial. Bias often arises from the datasets used to educate AI fashions, which might also mirror existing societal inequalities. Developers should proactively deal with this problem with the aid of the usage of various, balanced datasets and employing strategies like antagonistic debiasing to perceive and reduce discriminatory styles37. For example, if an AI machine is used for parole choices and indicates a bent to desire individuals from positive socioeconomic backgrounds, developers ought to retrain the system with impartial data to make sure equitable suggestions. Legal specialists and developers ought to additionally implement everyday bias assessments, using gear like counterfactual fairness

34 Nishith Desai Associates, Emerging Legal Issues with Use of Generative AI, (October 27, 2023),

https://www.nishithdesai.com/NewsDetails/10818. 35 Yogesh Tripathi et al., InSaAF: Incorporating Safety through Accuracy and Fairness | Are LLMs ready for the Indian Legal Domain?, (February 16, 2024), https://arxiv.org/abs/2402.10567. 36 Mathew Chacko, Aadya Misra, and Ajeeth Srinivas, Artificial Intelligence 2024 - India - Global Practice Guides, Chambers and Partners, (2024), https://practiceguides.chambers.com/practice-guides/artificialintelligence-2024/india. 37 Jasleen Kaur, The Rise of Artificial Intelligence and the Legal Challenges of Algorithmic Accountability in India, Lawful Legal, (January 19, 2025), https://lawfullegal.in/the-rise-of-artificial-intelligence-and-the-legalchallenges-of-algorithmic-accountability-in-india/.

testing to simulate how AI selections could trade if inappropriate factors, consisting of race or gender, had been altered. By prioritizing bias mitigation, the felony zone can uphold its dedication to justice and prevent AI from perpetuating or amplifying inequalities.

COPYRIGHT CHALLENGES:

One of the maximum urgent troubles in AI copyright regulation is the ownership and authorship of works created by means of AI. While the Indian Copyright Act in most cases emphasizes human authorship, the popularity of AI as a joint writer within the Suryast case suggests a potential shift in how the regulation may additionally evolve. The act of granting copyright to an AI gadget, alongside a human, marks an acknowledgment that AI may be more than only a tool inside the innovative process. However, this choice continues to be a unique instance, and the wider utility of this principle stays unsure38. The project turns into even more complicated while AI contributes to works with minimal human involvement.

For instance, if an AI gadget creates an artwork or composition largely without human enter, questions get up approximately whether the human consumer or the developer of the AI have to preserve the copyright. As AI keeps benefitting autonomy, those questions would require more nuanced criminal answers39.

CONCLUSION:

As artificial intelligence maintains to redefine legal landscapes, navigating the complexities of copyright possession and legal responsibility stays a vital task. The specific nature of AIgenerated outputs needs a paradigm shift in conventional highbrow property frameworks, requiring a delicate balance between fostering innovation and protecting human creators' rights. Legal doctrines must evolve to recognize AI's function in innovative techniques whilst making sure that authorship remains anchored in accountable human oversight.. Policymakers and prison practitioners must try to balance the need for innovation with the safety of highbrow belongings rights. Emerging legal doctrines and reforms ought to offer clarity through 38 Noel Naiju George, Adapting Indian Copyright Law to the Age of Artificial Intelligence: Recognizing AI as Authors Under the Copyright Act of 1957, Indian J.

Integrated Res. L. , Vol. IV, Issue III, at 810 (2024), available at https://ijirl. com/wp-content/uploads/2024/06/ADAPTING-INDIAN-COPYRIGHT-LAW-TO-THE-AGE-OF-

ARTIFICIAL-INTELLIGENCE-RECOGNIZING-AI-AS-AUTHORS-UNDER-THE-COPYRIGHT-ACT-OF-

1957.pdf.

39 Thushar V. Senan, Abey Augustine & Aswathy Krishnan, AI, Creativity, and Copyright Law in India:

Navigating the Boundaries of Originality and Authorship, International Journal of Law Management and Humanities, Vol. 6, Issue 3, 2023, at 2941, https://ijlmh.com/paper/ai-creativity-and-copyright-law-in-indianavigating-the-boundaries-of-originality-and-authorship/.

redefining authorship to include AI contributions even as safeguarding the rights of human creators. At the equal time, legal responsibility frameworks must evolve to make sure duty for harm because of AI, whether via direct operator legal responsibility or shared responsibility fashions involving AI developers, customers, and entities deploying these structures. In addition, the responsibility structure should be adapted to address the moral and legal implications of AI-controlled decisions. Either through direct operator responsibility, developers, users use shared liability models or the establishment of AI-specific rules, and ensure that responsibility is crucial. The legal system should estimate the risk associated with AI-borne materials, address bias problems, misinformation and potential injury problems.

A collaborative, interdisciplinary technique regarding policymakers, criminal practitioners, AI

technologists, ethicists, and enterprise stakeholders is crucial to crafting criminal frameworks that uphold justice while embracing technological development. Through continuous dialogue, regulatory refinements, and ethical issues, we can establish a sturdy AI-pushed prison device— one which safeguards highbrow belongings rights, guarantees fairness, and mitigates dangers without stifling innovation. By doing so, the criminal network can offer clarity and certainty in an era wherein synthetic intelligence reshapes not simplest creative expression but also the essential standards of authorship and responsibility.