ABSTRACT
The use of Artificial Intelligence technologies is increasing in businesses and law. This has created challenges. One big question is whether contracts made, negotiated or carried out with AI are legally valid in India. The Indian Contract Act of 1872 is the law for contracts in India. It was made at a time when there were no automated systems. So, it does not. Allow AI systems to be involved in making contracts. This paper looks at whether contracts made with AI are legally valid. It also checks if the agreement is enough when AI takes the place of humans and who is responsible when there are problems with these contracts. The paper uses the Indian Contract Act of 1872 the Information Technology Act of 2000 and constitutional principles to make its points. It says that Indian law is not against contracts made with AI. However, it needs changes to the law to solve the problems that make it unclear in this area. The paper ends with suggestions for changing the law and guidance for courts. This will make Indian law suitable for the age of Artificial Intelligence. The paper examines factors such as AI-generated contracts, the sufficiency of consent, the allocation of liability. The Indian law framework requires legislative intervention. The intervention will address the lacunae that presently imperil legal certainty in this domain. It will render the Indian legal framework fit for purpose, in the age of intelligence.
KEYWORDS
Artificial Intelligence, Contract Formation, Indian Contract Act, Consent, Legal Liability, Smart Contracts, Information Technology Act
I. INTRODUCTION
The intersection of intelligence and contract law is a big deal in India but not many people have studied it. Artificial intelligence is changing how contracts are made with machines helping to create, check and carry out agreements without humans. This is happening fast. The laws in India have not kept up with the changes. The Indian Contract Act of 1872[1] says that for an agreement to be a contract people must agree to it freely be able to make a contract have a reason for making the contract and want something that is allowed by law. When machines are involved in making contracts it raises some big questions. Who is responsible when a machine makes a contract? Can a machine really agree to something freely? If something goes wrong with a contract made by a machine, who’s liable. The person who made the machine the company that used it or the person who used it? The Information Technology Act of 2000[2] helped a bit by saying that contracts made electronically are valid. It was made for old-fashioned software not for smart machines that can make decisions on their own. There is a gap between the laws we have and how artificial intelligence is being used.
This paper is structured as follows: Part II explains how we did our research, Part III looks at what other people have written about this topic, Parts IV, to VIII look closely at parts of contract law like how contracts are made who can make a contract, who is responsible if something goes wrong and what the laws say, Part IX gives some ideas for how the law could be changed, Part X concludes.
II. RESEARCH METHODOLOGY
This paper uses a research method and it also uses comparison and analysis. The main things I look at are laws, court decisions and official reports. These are the things that I use to analyze. I also use things like academic articles, parts of books and policy documents to give more context and to criticize the main things.
I compare what is happening in the United Kingdom, the European Union and the United States. These places are ahead of us when it comes to rules about Artificial Intelligence and contracts. I look at the Law Commission of Indias Report No. 232 On Formation of a Contract by Electronic Means and the NITI Aayog National Strategy for Artificial Intelligence. These reports show what the Indian government is doing about Artificial Intelligence.
I do not use real life data. Go out and do research in the field. Instead, I carefully read the texts make conclusions and give my opinion on what the rules should be. Artificial Intelligence is what I am focusing on. The rules, about Artificial Intelligence and contracts are what I am analyzing.
III. REVIEW OF LITERATURE
People are just starting to look at how artificial intelligence generates contracts in India. There are many studies from other countries that can help us understand this topic. Ryan Abbotts idea of the ” computer” is very helpful in thinking about how the law should handle artificial intelligence. He says that our current laws are not good enough to deal with the decisions made by machines.[3]
In India, Divan and Sengupta have done a lot of work to figure out how artificial intelligence and the law interact[4]. They found that the law is having trouble keeping up with changes in technology especially when it comes to contracts, injuries and intellectual property. They think that the courts in India are not willing to change the law enough so the government should step in and make new laws.
There is a set of rules called the UNCITRAL Model Law on Electronic Commerce[5] that says contracts made electronically should be considered valid. This idea has been partly adopted in Indias Information Technology Act, 2000. However, these rules were made before artificial intelligence could make decisions on its own so they do not fully apply to intelligence systems.
The Law Commission of India looked at how electronic contracts are made and suggested changes, to the Indian Contract Act, 1872. Some of these changes have been made. They do not fully address the issues raised by artificial intelligence. This paper wants to explore these issues specifically looking at artificial intelligence generated contracts and how they affect the law.[6]
IV. LEGAL VALIDITY AND ENFORCEMENT OF AI-GENERATED CONTRACTS IN INDIA
4. AI-Generated Contracts: Definitional Contours
For this paper an AI-generated contract is a contract that is made accepted or agreed to by an intelligence system. This system works on its own. Does not need a person to check and approve each part of the contract. This means that AI-generated contracts can be anything from contracts where a computer just helps with writing to contracts that are completely made and signed by machines without any help.
There is a type of AI-generated contract called a contract. This is a computer program that is stored on a blockchain and automatically enforces its rules when certain conditions are met. Smart contracts have some issues because they cannot be changed and are not controlled by one person. But they also have some of the problems as other AI-generated contracts so we will consider them as part of the bigger group of AI-generated contracts, for this discussion.
4.2 Validity of AI-Generated Contracts under the Indian Contract Act, 1872
Section 10 of the Indian Contract Act, 1872 identifies the essential ingredients of a valid contract: free consent, competence of parties, lawful consideration, and a lawful object. Section 2(h) [7]defines a contract as an agreement enforceable by law. The question is whether an agreement generated by an AI system satisfies these criteria.
When we talk about offers and acceptance the law does not say much about offers and acceptance. The Information Technology Act of 2000 has a section called 10A that helps with this problem. This section says that just because a contract is made electronically it is not invalid. The Supreme Court made a decision in the case of Trimex International FZE Ltd. Versus Vedanta Aluminium Ltd.[8] The court said that contracts made through messages are just as valid as contracts made in person. This idea should also apply to contracts made by intelligence systems that are acting for a person. The artificial intelligence systems are working on behalf of a human so the contracts they make should be valid too. The law does not say anything against this so it should be okay. The contracts made by intelligence systems, like the ones acting for a human principal should be legally binding.
The more substantial doctrinal difficulty relates to the principal-agent framework. Classical contract law presupposes that an offer is made by a person natural or juristic who possesses legal personality and the intention to be legally bound. An AI system, being neither a natural person nor a juristic entity under Indian law, cannot itself be a party to a contract. The better analysis, therefore, is that the AI acts as an instrument or agent of the human or corporate entity that deploys it. The resulting contract is attributed to that deploying entity, much as the acts of a corporate employee bind the corporation. This attribution analysis is consistent with the general law of agency and with section 182 of the Indian Contract Act, which defines an agent as a person employed to do any act for another.
4.3 The Problem of Consent in AI-Mediated Contracting
Consent is the soul of contract. Section 13 of the Indian Contract Act, 1872 defines consent as the agreement of parties upon the same thing in the same sense.[9] Section 14 specifies that consent is “free” only when it is not caused by coercion, undue influence, fraud, misrepresentation, or mistake.[10] Both provisions presuppose human cognition, deliberation, and volition – capacities that an AI system does not possess in any philosophically robust sense.
In AI-generated contracts, the consent question bifurcates. First, there is the consent of the human party who programmes or deploys the AI: by configuring the AI to transact on defined parameters, the deployer may be said to have given a prospective, conditional consent to contracts falling within those parameters. This analysis draws on the concept of “general authority” in agency law, an agent with general authority to transact binds the principal within the scope of that authority. Second, there is the consent of the counterparty, who may or may not know that they are contracting with an AI system. Where a counterparty is unaware of the AI’s involvement, questions of mistake including, potentially, common mistake as to the nature of the contracting party – may arise.
The Restatement (Second) of Contracts identifies mutual assent as the cornerstone of contract formation.[11] Indian courts have similarly emphasised the requirement of a meeting of minds. Where one “mind” is an AI, it is unclear whether the mutuality requirement is satisfied. A purposive reading of the statute, attentive to the commercial necessity of giving effect to AI-mediated transactions, would support validity; a literalist reading might undermine it. The resolution of this ambiguity is ultimately a legislative task, though courts can and should adopt the purposive approach in the interim.
4.4 Contractual Capacity and the AI Actor
Section 11 of the Indian Contract Act 1872 says that to make a contract a person must be an adult have a mind and not be stopped from contracting by any law. A computer program does not fit these rules. It is not a person in the eyes of the law it does not have an age. Its working is not like a human mind.
This does not mean contracts made by computer programs are invalid. If we think of a computer program as a tool used by a person or a company then it can make contracts. Like a company can make contracts through its authorised representatives a company or a person can make contracts through a computer program. The important thing is that the person or company behind the computer program must be able to make contracts and must have told the program what it can do.
There is a problem with computer programs that can learn and do things on their own. If such a program makes a contract that it was not supposed to make the contract might be cancelled. This is similar to when someone acts for another person without their permission. The person who made the contract, with the computer program would then have to figure out if the company that used the program had made it seem like the program could make that contract. This would depend on what the company did that made the person think the program had permission.
4.5 Liability for Breach and Harm: The Allocation Problem
Liability is an issue when it comes to contracts made by artificial intelligence. There are least three people or groups who could be held responsible: the person who created the artificial intelligence system, the company that uses the intelligence system and the person who interacts with the intelligence system who is responsible depends on what happened and why. If the artificial intelligence system makes a mistake because it was designed poorly the creator of the system might be responsible. For example, if the artificial intelligence system always makes contracts with terms that’re not fair the creator might be liable.
The Information Technology Act of 2000 has rules about when a company can be held responsible for problems with data. These rules could also be used for problems with contracts made by intelligence systems.[12]
If the problem is with how the company uses the intelligence system, then the company is responsible. The Supreme Court has said that companies should be held to the contracts made by their intelligence systems as long as the contracts are what the company intended. This makes sense because a company is responsible for what its employees do.[13]
When it comes to consumers the Consumer Protection Act of 2019 gives them protection. If an artificial intelligence system makes a contract that’s misleading or unfair the company that used the system might be responsible.[14] However artificial intelligence systems can make many contracts that it is hard for individual consumers to get help. This means that the government needs to get involved to regulate these systems.
The artificial intelligence system is the key, to all of this. Artificial intelligence systems are making contracts and artificial intelligence systems are the problem. The people who make the intelligence systems and the companies that use them need to be careful. They need to make sure that the artificial intelligence systems are fair and do not make contracts.
4.6 Smart Contracts and Specific Performance
Smart contracts present a distinct enforcement challenge. Because they execute automatically upon the occurrence of specified conditions, they may perform obligations, including payment or transfer of assets, before a court has the opportunity to grant relief. The question of whether such automatic execution constitutes “specific performance” within the meaning of the Specific Relief Act, 1963, and whether courts can compel the reversal of a self-executed smart contract, remains doctrinally unsettled.
The Supreme Court’s decision in Bharat Sanchar Nigam Ltd. v. Motorola India Pvt. Ltd. affirmed that courts will give effect to the terms of commercial agreements as understood by reasonable commercial parties.[15] Applied to smart contracts, this suggests that courts would treat the self-executing mechanism as a contractual term to which parties consented, and would decline to reverse automatic execution absent fraud or fundamental breach. The practical consequence is that parties to smart contracts must ensure, at the contracting stage, that they incorporate appropriate error-correction and dispute-resolution mechanisms a task for which legal and technical expertise must collaborate.
4.7 Data Protection, Privacy, and AI Contracting
AI systems deployed for contracting purposes necessarily process large volumes of personal data. The Digital Personal Data Protection Act, 2023 imposes obligations on data fiduciaries, including entities that deploy AI for commercial purposes to process personal data only for specified, lawful purposes and with the consent of the data principal.[16] Where an AI contracting system collects or processes personal data beyond its specified purpose, for instance, by inferring sensitive personal attributes to tailor contract terms – the deploying entity may be in breach of its data protection obligations.
The intersection of data protection and contract law is particularly acute in the context of personalised or discriminatory contracting. An AI system trained on biased data may systematically offer less favourable contract terms to persons belonging to particular demographic groups. This raise concerns under both the Digital Personal Data Protection Act, 2023 and the constitutional guarantee of equality under Article 14 of the Constitution of India, though the application of fundamental rights to private contractual relationships is itself a contested doctrinal question.
4.8 Comparative Perspectives: The EU AI Act and Implications for India
The European Union made a law about Artificial Intelligence in 2024. This law is very detailed. It tells us how Artificial Intelligence systems should be controlled. The law looks at Artificial Intelligence systems.[17] Decides how risky they are. It then tells the people who make and use these systems what they have to do. They have to be honest about what these systems can do. They have to make sure people are in charge of them. This is especially important for systems that people use when they buy things.
Even though India does not have to follow this law it will probably still have an effect on India. This is because many Artificial Intelligence systems used in India are made by companies that have to follow the European Union’s law. India can learn from what the European Union’s doing and use that to make its own laws.
The NITI Aayog made a plan for Artificial Intelligence in India in 2018. This plan said that Artificial Intelligence is an opportunity for India and it proposed some basic ideas, for how to control it. The plan does not talk about contracts. It does say that people who make Artificial Intelligence systems have to be responsible and honest. It also says that citizens have to be protected. These are ideas that India can use to make more specific laws.[18]
India should look at what the European Union’s doing and use that to make its own laws. It should not just copy what the European Union is doing. It should make laws that work for India.
V. SUGGESTIONS
Based on the foregoing analysis, this paper advances the following normative recommendations for legislative and judicial action.
First, the Indian Contract Act of 1872 needs to be changed. We should make it clear that contracts made using Artificial Intelligence systems are legal long as they are made with the permission of a person or company in charge. We need a rule that says a contract is not invalid just because one or both parties used an Artificial Intelligence system to negotiate or sign it. This new rule should be similar to a rule in the Information Technology Act.
Second, we also need to make rules about Artificial Intelligence acting on behalf of people or companies. The law should say when a person or company is responsible, for what an Artificial Intelligence system does. We need to figure out what happens when an Artificial Intelligence system does something it was not supposed to do because it learned something. We should make it clear when a contract can be cancelled in these situations. The Indian Contract Act and Artificial Intelligence systems are important here. Artificial Intelligence systems are being used more and more to make contracts.
Third, mandatory disclosure requirements. Deployers of AI contracting systems should be required to disclose, at or before the point of contracting, that the agreement has been generated or negotiated by an AI system. This disclosure obligation would protect the consent of counterparties who may be unaware of the AI’s involvement, and would reduce the incidence of mistake as a ground for avoidance.
Fourth, a liability framework for AI contracting harms. A dedicated statutory regime – or, alternatively, an amendment to the Information Technology Act should allocate liability for harms arising from AI-generated contracts. The framework should distinguish between design defects (for which developers bear primary responsibility), deployment defects (for which deployers bear primary responsibility), and user-induced errors (for which users bear responsibility), while preserving the right of injured parties to proceed against any defendant in the chain.
Fifth, sector-specific regulation for high-stakes AI contracting. AI systems deployed in high-stakes contracting contexts including financial services, healthcare, employment, and insurance should be subject to sector-specific oversight by the relevant regulators (the Reserve Bank of India, the Insurance Regulatory and Development Authority, and so forth). Regulators should have the power to require human review of AI-generated contracts above defined value or complexity thresholds.
Sixth, we have training and the ability of institutions to handle things. Indian courts are going to have to make decisions about disputes that involve contracts made with the help of Artificial Intelligence. So, it is an idea to invest in training programmes for judges about Artificial Intelligence and technology law. We should also work on creating ways to resolve disputes like having people who are experts in technology help, with decisions and having panels of experts. This will really help make the decisions made by courts better and more consistent when it comes to Artificial Intelligence.
VI. CONCLUSION
The Indian legal system is not ready to handle contracts made by intelligence. The Indian Contract Act of 1872 is an important law that gives us the basic principles but it does not say anything about contracts made with the help of artificial intelligence. This creates a lot of uncertainty that’s bad for business and does not protect the people who are making the contracts.
This paper shows that it is possible to make artificial intelligence contracts work within the system by using some creative thinking and applying the principles of agency law. The contract is seen as belonging to the person or company that is using the intelligence. The consent is thought of as the person or company agreeing to the contract and the responsibility for any problems is given to the person or company that made the mistake. However, this is not a long-term solution because it relies too much on the judge’s decisions and does not protect the other parties in the contract who may not know that artificial intelligence is being used.
What we need is a law that specifically says artificial intelligence contracts are allowed sets out the rules for them to be valid requires transparency clearly says who is responsible if something goes wrong and has special rules for industries where a lot is at stake. India can be a leader in this area by using its experience with law its values of equality and dignity and its growing importance in the technology world. If India makes this law, it will answer some very important questions and show that it is ready to take part in the global economy that is driven by artificial intelligence.
The Indian Contract Act of 1872 is a law but it needs to be updated to include artificial intelligence contracts. Artificial intelligence contracts are the future. India needs to be ready. Artificial intelligence contracts need to be transparent. The rules need to be clear. India can make a law that will help the country participate in the global economy. Artificial intelligence contracts are a thing and India can be a leader, in this area.
Author: AVANIJA TYAGI
College/Institution: Amity Law school, Noida
Year/Course: LL. B (H), Final Year
Date: May 2026
[1] Indian Contract Act, 1872, sec 10 (India).
[2] Information Technology Act, 2000, sec 10A (India).
[3] Ryan Abbott, The Reasonable Computer: Disrupting the Paradigm of Tort Liability, 86 Geo. Wash. L. Rev. 1 (2018).
[4] Shyam Divan & Arghya Sengupta, Artificial Intelligence and the Law in India: Challenges and Opportunities (2022).
[5] UNCITRAL Model Law on Electronic Commerce, art. 11 (1996).
[6] Law Commission of India, Report No. 232: Formation of a Contract by Electronic Means (2009).
[7] Indian Contract Act, 1872, sec 2(h) (India).
[8] Trimex International FZE Ltd. v. Vedanta Aluminium Ltd., (2010) 3 SCC 1 (India).
[9]Indian Contract Act, 1872, sec 13 (India) (defining consent as parties agreeing upon the same thing in the same sense).
[10]Indian Contract Act, 1872, sec 14 (India).
[11]Restatement (Second) of Contracts sec 20 (Am. L. Inst. 1981) (mutual assent and meeting of minds).
[12] Information Technology Act, 2000, sec 43A (India) (corporate liability for data negligence).
[13] M/S Centrotrade Minerals & Metals Inc. v. Hindustan Copper Ltd., (2017) 2 SCC 228 (India) (party autonomy and binding nature of commercial agreements).
[14] Consumer Protection Act, 2019, sec 2(7) (India).
[15]Bharat Sanchar Nigam Ltd. v. Motorola India Pvt. Ltd., (2009) 2 SCC 337 (India).
[16]Digital Personal Data Protection Act, 2023 (India).
[17] European Parliament, Artificial Intelligence Act, Regulation (EU) 2024/1689 (2024).
[18] NITI Aayog, National Strategy for Artificial Intelligence (2018).