Reconceptualizing Partyhood In Contracts In The Age Of Artificial Intelligence - Can AI Be A ‘Party’ To A Contract
Author: LawSphereX
Category: Technology, Cyber & Privacy
Published:
Updated:
RECONCEPTUALIZING PARTYHOOD IN CONTRACTS IN
THE AGE OF ARTIFICIAL INTELLIGENCE: CAN AI BE A
‘PARTY’ TO A CONTRACT?
Navya Mehtani, B.B.A. LL.B. (Hons.), Faculty of Law, Delhi University, India
ABSTRACT:
In this era of modern technology, increasingly, AI systems are starting to independently generate contractual terms, acting as drafting assistants, being used to perform high-speed risk analysis and be used for compliance checks or simple renewals with minimal human intervention. While existing legal scholarships largely maintain that AI cannot be recognized as a contracting party due to its lack of legal personality, such conclusions tend to rely on traditional assumptions of contractual intent and capacity, that predate the logical decision- making. According to Saad Misbah and Inaya Imtiyaz, “determining liability whether it should fall on the AI developer or the AI itself, is complex due to a lack of clear legal guidelines, and it’s also complex if the contract operates across borders, identifying the appropriate legal jurisdiction”1.
This paper examines whether the conventional contract doctrine is fit and adequately equipped to deal with the contracts that are formed and performed through AI systems. It critically analyses the capability of AI to act as a party to a contract, by analysing the relation between intent, legal personality, and contractual capacity. The paper questions whether these elements must necessarily coexist in all contracting entities, and whether duties can be imposed on AI without it having the rights of a contracting party. Drawing upon the principles of contract law, agency, and technological realities, the paper explores whether alternative doctrinal models, like limited recognition of AI within contractual framework can address contemporary needs and realities in a better manner.
It concludes by highlighting the doctrinal gaps within the existing laws, and proposes a reconceptualization of contractual partyhood that answers the complexities, while preserving the morality and foundational principles of contractual accountability and laws.
1 Saad Misbah and Inaya Imtiyaz, “AI IN CONTRACT LAW: NAVIGATING LEGAL CHALLENGES AND
OPPORTUNITIES IN THE DIGITAL ERA” 2 LIJDLR 264 (2024).
Keywords: Artificial Intelligence, Contract Law, Contractual Partyhood, Legal Personality, Contractual Capacity, Autonomous Systems, Agency Theory.
INTRODUCTION:
Contract law is based on human agency, intentionality, and accountability. As written by Ágnes Juhász, “Contract law is the most dynamic part of civil law. This characteristic shows itself not only in the operation but in the development of contract law, which has been adapting itself for centuries to the needs raised by everyday life2. ” The foundational doctrines of offer, acceptance, consent, and capacity to contract assumes that the person entering into the contract has the cognitive ability to intend to contract, and the legal capacity to be bound by the obligations thereunder. These foundations are increasingly being challenged by the advent and evolution of Artificial Intelligence (AI) in commercial and contract law. In the past few decades, AI has revolutionised several industries, be it technology, healthcare, education or even the legal field.
However, severe dependence on AI as a tool for decisionmaking and interactions is a harmful and a generally fearful prospect, due to human involvement decreasing, and hence decreasing novel thoughts. Thus, in the recent years, AI systems have moved beyond passive tools that merely assist humans, to AI tools that are capable of autonomously generating contractual terms, negotiating the terms, selecting counterparties, etc. “As AI technologies continue to permeate the fabric of contractual relationships, it is imperative to explore the implications and ramifications of this integration3. ” Thus, as AI systems move closer to functional autonomy in decision making regarding contracts specifically, the question arises whether the existing contract doctrine can adequately explain or regulate their role within contractual relationships.
The prevailing rule is that AI, as of now, cannot be a contracting party due to its lack of legal personality, and under traditional doctrines, only legal persons (natural or artificial) possess the capacity to contract and be subject to the rights and obligations. An AI can and does not have legal identity as of yet, i. e. , it cannot sue or be sued. This means that right now, AI cannot be a party to a contract. The contract can just be between the human or corporate users of the AI 2 Ágnes Juhász, "The Applicability of Artificial Intelligence in Contractual Relationships" 9(1) Acta Univ. Sapientiae, Leg. Stud. 63 (2020). 3 Meenu, “The Impact of Artificial Intelligence on Contract Law: Challenges and Opportunities” 2(1) IJL 24 (2024).
system. 4 As a consequence, the actions of AI systems are typically attributed to human users or developers, through doctrines of agency or vicarious liability, which means that one person or entity is held legally responsible for a wrongful act committed by another, despite not directly causing it. This attribution based model sometimes raises significant conceptual and practical difficulties. The law regarding agency presupposes a principal who controls the agent and an agent who acts in accordance with what the principle instructs or intends. Autonomous AI systems challenges both these assumptions, as they might generate contractual outcomes that are not directly instructed or foreseeable by the human parties. In such cases, assigning contractual intent to a human principle may not be the right remedy. Existing doctrines offer little guidance as to how to address such gaps.
Currently, the debate remains between whether AI must remain a mere tool with no legal identity, or should it be elevated to the status of a full legal person. This binary view does not consider the more specific doctrinal questions concerning contract law. This paper argues that the growing autonomy of the AI necessitates a reconsideration of how contractual partyhood is seen, without resorting to an entire recognition of AI as a proper legal party. While it is argued that developers might try to escape liability if AI is granted identity, properly designed, such recognition need not displace human liability and accountability, but will act as a supplementary mechanism to strengthen contractual certainty.
Thus, the paper examines whether artificial intelligence can be meaningfully placed within contract law to guarantee contractual obligation without full legal personality, whether contractual partyhood can be understood in performance cantered terms, and whether the absence of moral and legal accountability necessarily precludes recognition within contracts. By answering these questions, the paper seeks to identify the doctrinal gaps created by AI mediated contracts and whether the existing legal categories remain adequate.
LITERATURE REVIEW:
1. Statutory Framework-
4 "4 Pillars of AI Contracts Validity: A 2025 Legal Guide", The Kanoon Advisors, Nov. 2, 2025, available at: https://thekanoonadvisors.com/4-pillars-of-ai-contracts-validity-a-2025-legalguide/#:~:text=This%20is%20the%20most%20significant,users%20is%20what%20determines%20competency. (last visited on Jan. 18, 2026).
• THE INDIAN CONTRACT ACT, 1872
The Indian Contact Act of 1872 is the primary statute governing contracts and their validity in India. Section 10 states that ‘agreements become contracts if they are made with free consent, by parties competent to contract, for lawful consideration and with a lawful object, and are not expressly declared void’5. The ICA specifies in Section 11 who is competent to contract. ‘A person must have attained the age of majority, must be of sound mind, and must not be disqualified by any other law.’6
• GENERAL CLAUSES ACT, 1897
As the ICA does not define ‘persons,’ section 3(42) of the General Clauses Act remains the standard interpretative tool for Central Acts that do not define ‘person’ themselves, stating that ‘persons’ includes both natural persons (humans) and legal/juridical persons (like companies and firms), thereby excluding AI7.
• INFORMATION TECHNOLOGY ACT, 2000
The IT Act recognises electronic records and digital signatures under Sections
4 and 3 respectively, thus facilitating electronic contracting. Further, Section
10A ensures that a contract cannot be denied enforceability solely because it was formed electronically8. However, the Act remains silent on the legal status of AI systems involved in contract formation.
• UNCITRAL MODEL LAW ON ELECTRONIC COMMERCE, 1996
It provides international guidance on electronic contracting and recognises the role of automated systems. Article 12 serves as the international standard for the legal recognition of automated transactions9. However, this too avoids
5 The Indian Contract Act, 1872 (Act 9 of 1872), s. 10.
6 The Indian Contract Act, 1872 (Act 9 of 1872), s. 11.
7 The General Clauses Act, 1897 (Act 10 of 1897), s. 3(42).
8 The Information Technology Act, 2000 (Act 21 of 2000), s. 10A.
9 UNCITRAL Model Law on Electronic Commerce, 1996, art. 12.
addressing the legal personality or contractual status of such systems.
2. Academic Literature-
The paper makes reference to several existing academic works addressing the topic and questioning it. “The Impact of Artificial Intelligence on Contract Law: Challenges and Opportunities” by Dr. Meenu talks about the transformative influence of artificial intelligence (AI) on contract law, examining the challenges and opportunities that arise from this symbiosis10. “The Applicability of Artificial Intelligence in Contractual Relationships” by Ágnes Juhász concludes by stating that ‘there is no doubt that law shall reckon with the massive expansion of intelligent contracts and that it shall answer in the future the difficult questions raised by them. One of these questions is if smart contracts can eventually replace traditional contracts, i. e. if smart contracts can appear as real alternatives of traditional contracts over time.
’11 Werbach and Cornell’s “Contracts Ex Machina” states that “while smart contracts offer novel possibilities and may alter the commercial world, they will not displace contract law. ”12 The paper makes reference to several other academic pieces, including old pieces like Tom Allen and Robin Widdleson’s “Can Computers Make Contracts,” which explores whether automated systems can legally make contracts13, and new ones like “AI in Contract Law: Navigating Legal Challenges and Opportunities in the Digital Era,” which states that “the challenge will be to ensure that the technological revolution aligns with the enduring principles of justice and equity and that the future of contract law lies
10 Supra note 3 at 2.
11 Supra note 2 at 2.
12 Kevin Werbach and Nicolas Cornell, "Contracts Ex Machina" 67 Duke Law J. 313 (2017), available
at: https://scholarship.law.duke.edu/dlj/vol67/iss2/2 (last visited on Jan. 19, 2026).
13 Tom Allen and Robin Widdison, "Can Computers Make Contracts?" 9(1) Harv. J.L. & Tech. 25
(1996), available at: https://jolt.law.harvard.edu/articles/pdf/v09/09HarvJLTech025.pdf (last visited on Jan. 19, 2026).
at the intersection of human wisdom and artificial intelligence”14.
RESEARCH OBJECTIVES:
This study is guided by the following objectives-
1. To examine conceptual foundations of contractual partyhood under traditional contract
law, with reference to the relationship between legal personality, capacity to contract, and enforceability.
2. To analyse whether contractual capacity can be doctrinally separated from full legal
personality, especially in the context of autonomous AI systems.
3. To examine whether Artificial Intelligence function as a legally recognised locus of
contractual obligation, without possessing the enforceable rights, i.e. can duties exist without full rights.
4. To evaluate the ability of existing agency- based models in addressing contracts
negotiated and executed by AI systems.
5. To explore alternative frameworks of contractual partyhood, and to propose a limited
and controlled model of contractual recognition for AI.
6. To identify and assess accountability gaps and liability risks arising from AI in
contracting.
RESEARCH QUESTIONS:
1. How does the increasing autonomy of AI systems challenge the traditional notions of
contractual partyhood, capacity to contract, and intent under Indian contract law?
2. Are the existing doctrinal frameworks, particularly regarding agency and legal
personality, adequate to address accountability and enforceability in AI contracting?
3. Can contract law accommodate a limited and controlled form of recognition for
artificial intelligence that enhances accountability without conferring full legal
14 Supra note 1 at 1.
personality?
RESEARCH GAPS:
Despite there being a growing number of papers on artificial intelligence and contract law, significant doctrinal gaps remain in how contract law conceptualises contractual partyhood, capacity to contract, and accountability in AI done transactions.
1. Existing scholarship largely assumes that contractual capacity flows from legal
personality, and there is limited doctrinal examination of whether capacity to be bound by contractual obligations can exist independently of full legal personhood, particularly in context of autonomous AI systems.
2. Current literature presumes that contractual partyhood requires a balance of rights and
obligations. There is minimal to no existing literature on the possibility of AI functioning as a locus or place of contractual obligation without possessing the complete contractual rights, challenging a fundamental theory under contract law.
3. Limited and functional models of partyhood, particularly where AI autonomously
determines and executes contractual performance remains unexplored.
4. There is insufficient analysis of whether these models adequately capture accountability
and enforceability concerns in autonomous AI contracts. Thus, there are unaddressed accountability and liability gaps.
5. There is no consideration as of yet as to whether AI could be recognised as a temporary
or event specific contracting party limited to contract formation.
HYPOTHESIS:
“Traditional contract law, which treats legal personality as a prerequisite for capacity to contract and partyhood, is not enough to address AI’s role in contractual activities; a limited and controlled form of contractual recognition of AI can enhance accountability and contractual relevance without there having to be full legal personhood granted.” This hypothesis is based on the assumption that increasing AI autonomy has exposed limitations within existing contractual statutes. By assuming that only legal persons possess
contractual capacity, current frameworks risk there being no accountability when the autonomous systems independently negotiate or perform contractual obligations.
RESEARCH METHODOLOGY:
The study adopts a doctrinal research methodology, focusing on the systematic analysis and interpretation of legal principles governing contractual partyhood, capacity, intent, and accountability. The research examines statutory provisions, judicial interpretations, and established doctrines withing contract law to assess their adequacy in addressing AI in contracting. Important statutory instruments, like the Indian Contract Act, 1872, the IT Act, 2000, etc. , were analysed to identify the assumptions regarding contracting parties, and to evaluate the applicability of those assumptions and provisions in the acts on autonomous AI systems. The study engages critically with the existing academic literature on AI, smart contracts, legal personality, and autonomous systems to figure out the prevailing approaches and to identify the gaps.
Relevant judicial decisions relation to capacity to contract, agency, AI in contracts, and electronic or e-contracts are examined to understand how the courts have historically approached non- traditional ways of contracting. The research is confined to private law and does not extend to tort or criminal law. The analysis is theoretical and doctrinal in nature and does not involve empirical data or technical evaluation of artificial intelligence systems.
ANALYSIS:
General Historical Context of Contract Law: Contract law has significantly evolved over time, and it continues to develop as the society and the economy changes, technology advances and environment transforms. The evolution of contract law reflects a transition from rigid, formalistic rituals in ancient societies to a modern system based on mutual agreement and economic utility.
Ancient Roots- In ancient times, contract law was basic, and was based on oral agreements and customs that were present in the society. ‘The Roman law laid the groundwork for many principles of modern contract law, and introduced the concept of quid pro quo (something for something) and the maxim pacta sunt servanda (agreements must be kept)’15. As for India during the entire ancient and medieval periods, there was no general code covering contracts. Contractual principles were derived from religious texts like the Vedas, Smritis, and Arthashastra. Early laws emphasized free consent (Samanvaya) and defined specific conditions under which contracts (such as those made in secret or under intoxication) were void16.
• 18th to 19th centuries- There was a rise of the concept of ‘freedom to contract,’ and in India, ‘the English Law was applied in the towns of Madras, Bombay and Calcutta under the Charter of 1726, and enforceability was granted to promises with exceptions’17. • The Indian Contract Act (ICA), 1872- The ICA was finally drafted by the third Indian Law Commission in 1861, finally coming into effect in 1872, compiling all the laws related to contracts in India. In the amendments that followed, sections 76 to 123 dealing with the sales of goods, and 239 to 266 dealing with partnership were repealed and separate legislations were enacted called Sales of Goods Act 1930, and the Indian Partnership Act 1932 respectively.
15 Serhii Floreskul and Violetta Loseva, "The evolution of contract law", Avitar Blog, available
at: https://avitar.legal/blog/the-evolution-of-contractlaw#:~:text=In%20ancient%20times%2C%20contract%20law,the%20development%20of%20merchant%20law s (last visited on Jan. 23, 2026).
16 Sujoy Paul, "Historical Background Of Indian Contract Act 1872", Legal Service India, available
at: https://www.legalserviceindia.com/legal/article-7749-historical-background-of-indian-contract-act- 1872.html (last visited on Jan. 23, 2026).
17 Supra note 16 at 7.
Contemporary shift and Technological Evolution- With the advent of technological development in the world, it became important to have a statute governing the related laws and rules. Provisions like the Information Technology Act, 2000 recognize electronic signatures and econtracts as legally valid in India. Contractual Partyhood and Capacity: Section 2(h) of the Indian Contract Act, 1872, defines a contract as ‘an agreement enforceable by law’18. For this enforceability, agreements must be ‘made with free consent, by parties competent to contract, for lawful consideration and with a lawful object, and should not be expressly declared void’ as per section 1019, and the parties ‘must have attained the age of majority (18 years of age or 21 if a guardian is court-appointed), must be of sound mind, and must not be disqualified by any other law (e. g.
, alien enemies, convicts while serving their sentence, etc. )’ as per section 11 of ICA20. Contract law has thus historically treated contractual partyhood as contingent upon legal personality. Though the Act does not define “person,” the general interpretation remains that section 3(42) of the General Clauses Act shall act the standard interpretative tool for Central Acts that do not define "person" themselves, stating that ‘persons’ includes both natural persons (humans) and legal/juridical persons (like companies and firms), thereby excluding Artificial Intelligence as a person who can enter into and perform a contract21. In the case Mohori Bibee v. Dharmodas Ghose (1903), the Privy Council held that a contract entered into with a minor is void ab initio, i. e. no contract existed from the beginning, due to lack of contractual capacity22.
Since then, the Courts have clarified over time that a contract entered into with a minor for the minor’s benefit would be treated as a valid one. This decision highlighted that capacity was not a factual ability to transact, but a legal construct conferred by law. Further, under the Information Technology Act, 2000, electronic signatures and automated
18 The Indian Contract Act, 1872 (Act 9 of 1872), s.2(h).
19 Supra note 5 at 3.
20 Supra note 6 at 3.
21 Supra note 7 at 3.
22 Mohori Bibee v. Dharmodas Ghose, (1903) 30 IA 114.
"smart contracts" are recognized as valid forms of agreement, effectively allowing digital identities to act as agents for parties. Smart contracts are defined as self-executing digital agreements with terms written directly into computer code and stored on a decentralised blockchain network. While smart contracts are legally accepted in India, and liabilities are clearly defined, the same is not the case for autonomous AI systems in contracts. While smart contracts follow rigid, predefined "if-then" rules where the same input always produces the same output, the essence of AI systems is intelligent estimation. AI systems do not fall into the category of ‘persons,’ and hence cannot legally hold the capacity to contract. The law presupposes human actors, or artificial persons like companies, as holding the ability to contract, and AI does not have that kind of legal personality.
Hence, AI as of now cannot be treated as a party to a contract. Attribution and Agency: Judicial Comfort and its Limits- The courts have traditionally dealt and resolved cases involving non-human instruments through attribution. In Trimex International FZE Ltd. v. Vedanta Aluminium Ltd. , the Supreme Court of India upheld the validity of contracts concluded via electronic communication, affirming that technological mediation does not negate intent where human consent can be inferred23. Under Indian law, particularly the Information Technology Act, 2000 (Section 11), an electronic record is attributed to the "originator" if it was sent by: • The originator themselves. • A person authorized to act on behalf of the originator.
An information system programmed by or on behalf of the originator to operate automatically24. Thus, as of now, the deployer is legally responsible for the "choices" the AI makes within its programmed parameters. Article 12 of the UNCITRAL Model Law on Electronic Commerce establishes the principle of acknowledgment of receipt, and validates contracts formed by automated message systems,
23 Trimex International FZE Ltd. v. Vedanta Aluminium Ltd., (2010) 3 SCC 1.
24 The Information Technology Act, 2000 (Act 21 of 2000), s. 11.
regardless of human review at the time of formation25. These frameworks rely on the agency logic, that the machine works at the will of the master. However, these again do not consider the liability in cases of autonomous systems, especially in the era of technological evolution, where holding these systems accountable and liable is of the utmost importance. In Syed Abdul Khader v. Rami Reddy, the Supreme Court reiterated that an agent’s acts bind the principal only when performed withing the scope of authority26. Thus, the position of autonomous AI systems where ‘scope of authority’ might not be clearly defined, where the AI might generate terms, or execute the contract beyond direct human instruction, continued reliance on attribution becomes a risk. Contractual Capacity Without Legal Personality- A core research question addressed by the paper is whether contractual capacity can exist
without full legal personality. Indian Contract Law treats separate legal personality as a threshold for being capable to enter into a contract. Not only AI systems, but other entities, for example Hindu Undivided Families, are not capable to enter into contracts, as they have no separate legal personality. However, it is visible through cases that law can recognise functional capacity without full personality. In the case of Shiromani Gurudwara Parbandhak Committee (SGPC) v. Som Nath Dass, it was established that the Guru Granth Sahib is a juristic person capable of owning property and entering legal transactions for limited purposes27. Applying this logic, contractual capacity may be reconceptualised as context-specific legal attribution. As AI systems become more autonomous (e. g.
, driverless cars or high-frequency trading bots), it becomes difficult to trace a single "harmful" action back to a human developer's intent. Granting AI limited personhood would allow it to be sued directly or hold its own liability insurance. Indian courts have already recognized deities, idols, and even rivers (though later stayed) as juristic persons to protect specific interests. Proponents argue AI deserves similar status to ensure responsible governance in areas like healthcare and finance. 28
25 UNCITRAL Model Law on Electronic Commerce, 1996, art. 12.
26 Syed Abdul Khader v. Rami Reddy, (1979) 2 SCC 601.
27 Shiromani Gurdwara Parbandhak Committee v. Som Nath Dass, (2000) 2 SCC 186.
28Kushagre Vats, “Beyond Human Hands- Rethinking Legal Status And Responsibility For AI In India” IJLSSS 3(4) 12.
Duties Without Rights: AI as the Locus for Contractual Obligation- Another central research concern is whether contractual duties can exist without corresponding rights. The concept of ‘Duties Without Rights’ represents a functional legal shift where AI is treated as a locus for obligation, i. e. , an entity that can bear duties and perform contractual tasks, without being granted the corresponding legal rights (such as the right to property, privacy, or constitutional protection) typically afforded to legal persons. Under the 2026 techno-legal frameworks, the ‘duty’ of the AI is legally attributed to the human/corporate controller. If the AI ‘fails’ its duty, the controller is strictly liable, even if the AI's decision-making process was autonomous. Some scholars argue that high-stakes AI (in healthcare or finance) should be coded with "fiduciary duties.
" This means the AI is legally mandated to act in the best interest of the human user, creating a one-way street of legal obligation. Hence, some arguments also stand that AI as a system should also be liable, not only their human controllers. The concept, applied to AI, recognising it as a non-human locus of obligation, allows contract law to avoid the ‘moral panic’ of granting AI civil rights while solving the ‘accountability gap. ’ This avoids the normative difficulties of rights attribution, while addressing accountability gaps created by autonomous performance. Why Limited Legal Personality Matters in AI Contracting? – The attribution of limited legal personality to AI is not an attempt to anthropomorphise technology (i. e. attribute human traits to technology), but a legal response to functional autonomous systems.
Granting AI limited personality is considered a pragmatic necessity to bridge the accountability gap in automated contracting without equating machines to humans. Organizations increasingly face the "AI accountability gap," where autonomous systems make high-speed decisions that humans cannot supervise in real-time. Limited personality allows for the direct attribution of specific legal acts to the AI system itself, rather than leaving victims to untangle a complex web of liability between developers, vendors, and users. 29 Furthermore, the prevailing legal approach treating AI ass a tool and attributing liability to
29 Supra note 28 at 10.
programmers and developers offers initial accountability, and becomes inadequate as AI systems exhibit adaptive learning and decision making, especially in sectors of healthcare, banking and finance, and judiciary. Holding programmers liable is viable, but also risks imposing responsibility for decisions that emerge from post-deployment learning, which is why, it becomes important to hold AI responsible along with the developers. Also, similar to a corporation, limited personality would enable AI systems to hold dedicated insurance policies or asset pools. This ensures that if a contract is breached, there is a prefunded mechanism for compensation, protecting human creators from unlimited personal liability for unpredictable AI behaviour. Finally, traditional punishments (like prison) are inapplicable to AI.
Limited personality allows for ‘penalties’ tailored to software, such as temporary disablement, mandatory code modification (debugging), or permanent deletion (incapacitation). Limitations and Challenges- While there are several arguments in favour of granting legal personhood to AI, there are also several arguments against the idea. The first and foremost remains the fear that assigning liabilities to AI would lead to a demand for rights, as managing the concept of ‘duties without rights,’ though possible, is difficult. Second remains the fact that unlike humans, AI lacks consciousness and moral responsibility. Legal personhood without moral agency may be hollow. 30 On a global platform as well, no institution has recognised the possibility of limited legal capacity for AI (like the UNCITRAL Model Law), which makes the position all the more unstable and confusing.
In the US, AI is treated as a product and has product as well as tort liability, and in cases like Saudi Arabia, it grants symbolic citizenship to Sophia, but no enforceable rights or duties. 31 30 Anamika Chaudhary, “Artificial Intelligence And Legal Personhood: A Critical Analysis In The Indian Context” IJLSSS 3(4) 66.
31 Supra note 30 at 11.
Limited Legal Personality as a Middle Path- Recognising limited legal personality for AI offers a middle path between full personhood and total objectification. Limited legal personality would allow the law to hold the developer liable for their faults but protect people from unlimited liability by holding the AI liable for things done outside the authority granted. By anchoring contractual obligations to the AI system’s autonomous conduct, courts can more accurately assess breach, causation, and risk allocation, while still ensuring human accountability. Critically, legal personality does not absolve developers or deployers of responsibility, it actually restructures responsibility in a way to mirror technological responsibilities. It enables clearer contractual drafting, better insurance mechanisms, and more predictable enforcement outcomes.
PROPOSED SOLUTIONS:
Thus, the paper advances the following proposed suggestions to reconceptualise contract law’s response to autonomous artificial intelligence systems, while preserving its foundational principles-
1. There should be a middle ground approach that allows liability for both the developers
and the AI. Contract Law should acknowledge a limited and purpose specific form of contractual identity for highly autonomous AI systems.
2. There should also be moves for introducing specific legislations to deal with AI
accountability and liability.
3. Parties engaging with AI mediated contracting should be required to explicitly allocate
risk, responsibility, and decision-making authority within contractual terms. Standard form clauses recognising AI accountability can help mitigate uncertainty and confusion.
4. National and international commissions to help deal with legal and ethical issues with
existing AI systems.
CONCLUSION:
The question revolving whether AI can and should be treated as a party to a contract is both
complicated and fascinating. As of now, autonomous AI systems do not have the kind of legal personality to enter and be treated as parties to contracts. However, the current attribution-based models, attributing liabilities to developers remains insufficient as AI develops and evolves. Thus, the paper strives to propose a middle path approach of granting limited liability and limited partyhood to AI systems. Rather than there being full legal personhood for artificial intelligence, this paper argues for a more restrained and functional approach. By recognising a limited, controlled form of contractual identity for AI, contract law can preserve doctrinal coherence while responding effectively to technological change.
This framework allows duties to exist without corresponding rights, allows performance-based understandings of partyhood, and accommodates temporally bounded recognition, each of which challenges, but does not undermine, foundational contract principles.