Published On: July 28, 2026
Authored By: Snigdho Dhar
University of Engineering and Management, Kolkata
Abstract
Artificial Intelligence has become one of the key technologies of the twenty-first century. It is used in fields like healthcare, finance, education, transportation and law enforcement. Artificial Intelligence helps in making processes more efficient and supports economic growth. However, it also raises complicated legal issues. These include questions about who is responsible when things go wrong, privacy concerns and whether laws are being followed. The traditional legal system was built around human decision-making and often struggles with problems caused by machines that can work on their own. In India, many people are using Artificial Intelligence, but the laws that control it are not yet in place. This article examines the idea of responsibility in the context of Artificial Intelligence, analyzes the challenges it brings and reviews the current laws in India relating to Artificial Intelligence. It also looks at how other countries are handling these issues. The article suggests that India needs to create a system that encourages innovation while ensuring responsibility, transparency and protection of basic rights.
Introduction
Artificial Intelligence refers to machines and computer systems that can perform tasks that usually require intelligence, such as learning, reasoning and problem-solving. Recent developments in machine learning and natural language processing have made Artificial Intelligence systems more capable.[1] Artificial Intelligence is now a part of daily life, from self-driving cars to medical tools and virtual assistants. As people rely more on Artificial Intelligence, there are concerns about who is responsible when these systems cause harm. For example, if a self-driving car has an accident, who is to blame? What if an Artificial Intelligence algorithm treats certain groups unfairly? Can an Artificial Intelligence system itself be held responsible for its actions? These questions challenge existing legal concepts that are based on human actions and intentions. Since Artificial Intelligence systems can function with varying levels of independence, it is hard to assign responsibility. India currently does not have laws for Artificial Intelligence, which makes the situation even more complex.[2]
Understanding Artificial Intelligence and Legal Liability
Legal liability occurs when someone or something is legally responsible for causing harm to another person. Liability can come from tort law, criminal law, contracts or regulations. When it comes to Artificial Intelligence, it is difficult to determine who is responsible because these systems often operate independently and can make decisions without human involvement. Unlike traditional software, advanced Artificial Intelligence systems can learn from data and change their behavior over time.[3] Many people are involved in the Artificial Intelligence ecosystem, including developers, manufacturers, organizations, end-users and data providers. When harm occurs, it is challenging to determine who should be held accountable.
Legal Challenges Posed by Artificial Intelligence
1. The Issue of Accountability
One of the key concerns is the gap in accountability caused by machines that can operate on their own. Traditional legal principles assume that actions can be traced back to human beings. However, Artificial Intelligence systems may produce results that were not intended or expected by their creators. For example, an Artificial Intelligence-based hiring tool may unintentionally discriminate against women or minority candidates because it learned from past data. Even though no individual may have planned this discrimination, it can still cause significant harm. This creates uncertainty about who should be held responsible.
2. Algorithmic Bias and Discrimination
Artificial Intelligence systems are only as good as the data they are trained on. If the data contains biases, the Artificial Intelligence models can replicate or even worsen these biases. There have been cases of discrimination reported globally in areas such as hiring, credit scoring and criminal justice. In India, algorithmic bias could affect certain communities more than others, raising issues about equality and protection against discrimination under Articles 14 and 21 of the Constitution.[5][6]
3. Privacy and Data Protection Concerns
Artificial Intelligence systems rely heavily on large amounts of data. Personal information collected from individuals may be used for profiling, surveillance and predicting behavior. The right to privacy is recognized in India as a fundamental right, as affirmed in Justice K.S. Puttaswamy (Retd.) v. Union of India.[7] Artificial Intelligence-driven data processing must follow these standards. The Digital Personal Data Protection Act, 2023 is a step towards regulating data processing, but specific rules addressing Artificial Intelligence-related risks are still limited.[8]
4. Autonomous Vehicles and Physical Harm
Autonomous vehicles highlight the complexity of Artificial Intelligence liability. If an Artificial Intelligence-controlled vehicle causes an accident, multiple parties may be held responsible, including the manufacturer, the software developer, the vehicle owner and the person monitoring the system. Traditional principles of negligence might not be enough to handle situations where machines make decisions.
5. Intellectual Property Challenges
Generative Artificial Intelligence systems can create text, music, art and other creative outputs. This raises questions about who owns Artificial Intelligence-generated content, whether there is copyright infringement during the training process and how to protect Artificial Intelligence-made inventions.[20] Existing intellectual property laws typically assume human authorship, which creates uncertainty about the legal status of Artificial Intelligence-generated work.
Existing Legal Framework in India
India currently does not have a specific law regulating Artificial Intelligence. However, several existing legal frameworks may apply.
A. Law of Torts
The law of torts provides remedies for wrongs that cause harm or loss. Victims affected by Artificial Intelligence systems may seek compensation under negligence principles. To prove negligence, a claimant must show a duty of care, breach of that duty, causation and damages. But proving negligence becomes difficult when Artificial Intelligence systems operate autonomously and their decision-making processes are not clear.
B. Consumer Protection Act
The Consumer Protection Act, 2019 offers remedies for faulty products and services.[9] If an Artificial Intelligence-enabled product causes harm due to defects, the manufacturer or service provider might be held responsible under product liability provisions.
C. Information Technology Act
The Information Technology Act, 2000 governs electronic transactions and cyber-related crimes.[10] Although it was enacted before modern Artificial Intelligence technologies existed, some of its provisions on cybersecurity, data protection and intermediary liability could be relevant in cases involving Artificial Intelligence.
D. Digital Personal Data Protection Act
The Digital Personal Data Protection Act, 2023 sets rules about how data is handled and what duties data controllers must follow. Artificial Intelligence creators and companies using AI systems often process large amounts of personal information, making compliance with these data protection rules essential.
E. Constitutional Safeguards
Constitutional principles also help constrain AI technologies. Any AI use by government bodies must meet constitutional standards, especially fairness and proportionality.
Comparative Approaches to AI Regulation
European Union
The European Union is leading the way in AI regulation with its Artificial Intelligence Act.[13] This law uses a risk-based approach, dividing AI systems into categories: unacceptable risk, high risk, limited risk, and minimal risk. High-risk systems must follow strict rules, such as transparency, record-keeping, and human supervision requirements.
United States
The United States has taken a different path, focusing on encouraging innovation in specific sectors. Instead of a single federal AI law, multiple agencies share responsibility for regulating AI within their respective domains.
China
China has started to set rules for recommendation algorithms and generative AI. Its approach emphasizes government supervision, transparency, and national security.
The Debate on AI Legal Personality
Some experts suggest that AI systems should be granted a form of legal status. Supporters argue that as AI systems become more independent, granting them legal status could make it easier to assign responsibility. But there are strong disagreements, since AI systems do not possess consciousness, moral judgment, or legal interests of their own.
Need for an AI Regulatory Framework in India
India is one of the world’s largest digital economies and is actively promoting AI innovation. However, regulatory uncertainty remains a significant problem. A dedicated AI framework should include principles like transparency, accountability, fairness, and human oversight. Organizations using AI should disclose when automated decisions are being made, and people affected by these decisions should receive explanations about how their rights are impacted. Developers, users, and operators must all bear responsibility for harms caused by AI, and effective remedies should be available to victims. AI systems should be regularly audited to identify and prevent unfair outcomes, and independent oversight can help ensure fairness.
Important decisions that affect life, freedom, healthcare, employment, and access to justice should not be made entirely by AI systems acting without human supervision; meaningful human oversight should always be required. Data governance is equally important — data must be collected, processed, stored, and shared lawfully, with data protection treated as a core component of AI regulation, not an afterthought.
Not all AI systems carry the same level of risk. India should therefore adopt a risk-based model, similar to the European Union’s, under which stricter rules apply to high-risk AI applications while greater flexibility is preserved for low-risk innovation.
To address these issues, the following measures are recommended:
1. Create AI laws that specifically address accountability and liability.
2. Set up a dedicated AI regulatory authority.
3. Conduct impact assessments for high-risk AI systems.
4. Require transparency and explainability standards for AI systems.
5. Strengthen consumer protection for harms caused by AI.
6. Develop sector-specific guidelines for fields like healthcare, finance, education, and law enforcement.
7. Encourage responsible AI development through industry standards and professional codes.
8. Promote international cooperation on AI governance.
9. Improve regulatory capacity to handle disputes arising from new technologies.
10. Ensure AI regulation aligns with constitutional values and human rights principles.
Conclusion
In conclusion, AI offers significant opportunities for economic growth, technological progress, and social development. However, its widespread use brings complex legal questions about liability, accountability, privacy, discrimination, and governance. Traditional legal systems were not built to handle problems arising from AI systems capable of making decisions on their own. Currently, India relies on a mix of tort law, consumer protection law, data protection rules, and constitutional safeguards to address AI-related issues, but these tools do not offer sufficient protection against the unique challenges posed by modern AI technologies. What is needed is a forward-looking legal system that supports innovation while ensuring accountability — one that guarantees technology does not compromise rights, public safety, or constitutional values. By adopting a risk-based approach grounded in transparency, human oversight, and strong accountability, India can benefit from AI while protecting justice, equality, and the rule of law.
References
[1] Stuart Russell and Peter Norvig, Artificial Intelligence: A Modern Approach (4th edn, Pearson 2021).
[2] Ryan Calo, ‘Artificial Intelligence Policy: A Primer and Roadmap’ (2017) 51(2) UC Davis Law Review 399.
[3] Harry Surden, ‘Artificial Intelligence and Law: An Overview’ (2019) 35 Georgia State University Law Review 1305.
[4] Ryan Abbott, The Reasonable Robot: Artificial Intelligence and the Law (Cambridge University Press 2020).
[5] Constitution of India, art. 14.
[6] Constitution of India, art. 21.
[7] Justice K.S. Puttaswamy (Retd.) v Union of India, (2017) 10 SCC 1.
[8] Digital Personal Data Protection Act, 2023 (Act No. 22 of 2023).
[9] Consumer Protection Act, 2019 (Act No. 35 of 2019), ss. 82–87.
[10] Information Technology Act, 2000 (Act No. 21 of 2000).
[11] NITI Aayog, National Strategy for Artificial Intelligence: #AIforAll (2018).
[12] Organisation for Economic Co-operation and Development (OECD), OECD Principles on Artificial Intelligence (2019).
[13] European Parliament and Council, Regulation Laying Down Harmonised Rules on Artificial Intelligence (Artificial Intelligence Act), Regulation (EU) 2024/1689.
[14] World Economic Forum, A Framework for the Development, Deployment and Evaluation of Trustworthy Artificial Intelligence (2022).
[15] Nick Bostrom, Superintelligence: Paths, Dangers, Strategies (Oxford University Press 2014).
[16] European Commission, ‘Artificial Intelligence Act’ (artificialintelligenceact.eu) accessed 21 June 2026.
[17] OECD, ‘OECD AI Policy Observatory’ (oecd.ai) accessed 21 June 2026.
[18] NITI Aayog, ‘Responsible AI for All: Approach Document for India’ (2021).
[19] World Economic Forum, ‘Global Technology Governance Report 2021’.
[20] Ryan Abbott, ‘I Think, Therefore I Invent: Creative Computers and the Future of Patent Law’ (2016) 57 Boston College Law Review 1079.
