Abstract
The rapid advancement of Artificial Intelligence (AI) has challenged traditional legal concepts of personhood. As AI systems are increasingly performing autonomous functions in areas such as finance, healthcare, transportation and governance, fundamental questions arise regarding their legal status. With this, the adequacy of existing legal frameworks also comes under scrutiny in addressing AI’s liabilities, rights and responsibilities in associated with its actions. This paper examines the concept of AI legal personality through the lens of classical jurisprudential theories of legal personality, namely the Fiction Theory, Concession Theory, Real Entity Theory, Purpose Theory and Bracket Theory. Under these theories the paper assesses whether AI can be accommodated within existing jurisprudential understandings of legal personhood or not. The paper further argues that AI does not satisfy traditional requirements of personhood because it lacks consciousness, moral agency, autonomous will or legally cognizable interests. Nevertheless, legal personality has historically been extended to non-human entities, including such as corporations, foundations, rivers and religious institutions for functional and regulatory purposes. The paper concludes that while AI should not be equated with natural persons or granted human rights, a carefully designed framework of limited legal personality may provide a pragmatic solution to the emerging legal challenges in the age of artificial intelligence.
Keywords:Artificial Intelligence (AI), Legal Personality, Juristic Person, Natural Person, Jurisprudence, Classical Theories of Legal Personality.
- Introduction
Artificial Intelligence (AI) has emerged as one of the most transformative technologies of the twenty-first century. Rapid advancement in machine learning, deep learning and natural language processing have enabled AI systems to perform tasks that were once considered exclusive to humans. This concept originated at the Dartmouth Workshop in 1956 and has evolved from simple rule-based systems to sophisticated deep learning models, emerging as one of the most influential technologies of the modern era.[1] It is now widely used across various sectors, including chatbots, search engines, smart technologies, healthcare, biotechnology and scientific research, etc. AI is generally classified into Artificial Narrow Intelligence (ANI), Artificial General Intelligence (AGI) and Artificial Super intelligence (ASI). Currently, all existing AI systems fall within the category of Artificial Narrow Intelligence (ANI) and are designed to perform specific tasks.[2]
As AI systems become increasingly sophisticated and capable of acting with greater autonomy, an important question arises: Should AI be recognized as a legal person? The concept of legal personality occupies a central position in jurisprudence, as it determines whether an entity is capable of possessing rights, bearing duties, owning property, entering into legal relations and being held accountable under the law or not. Historically, legal personality has not been limited to natural persons. Various non-human entities, including corporations,[3] Hindu idols,[4] the Holy Guru Granth Sahib ji[5] and in some jurisdictions, even natural objects such as rivers,[6]have been granted legal personality to achieve specific legal and policy objectives. This development suggests that legal personality is not merely a reflection of human characteristics but also a legal construct designed to serve practical and regulatory purposes. Consequently, the debate concerning AI legal personality has gained increasing attention among legal scholars, policymakers and international institutions,[7]
AI’s growing autonomy and decision-making capabilities have generated significant benefits while simultaneously raising complex legal, ethical and regulatory concerns. Its increasing integration into social and economic activities has revealed the limitations of existing legal frameworks. However, highly autonomous AI systems challenge conventional legal concepts relating to responsibility, liability, accountability and personhood. While AI offers significant economic and social benefits but also raises concerns regarding privacy, human autonomy and fundamental rights.[8] Proponents argue that granting a limited form of legal personality to advanced AI systems could facilitate the allocation of liability, regulatory oversight and effective governance.[9] Critics, however, contend that AI lacks consciousness, moral agency, free will and legally cognizable interests, thereby making the attribution of legal personality conceptually problematic.[10]Its rapid development has created regulatory challenges, highlighting the need for effective legal frameworks to address emerging risks and protect individuals from potential harm.[11] Consequently, the question of AI personhood requires a careful examination of the jurisprudential foundations of legal personality.
This paper seeks to reassess the classical theories of legal personality in the context of contemporary technological developments. It analyses the Fiction Theory, Concession Theory, Real Entity Theory, Aggregate Theory, Purpose Theory organic Theory, Normative Theory, Interest Theory and Will Theory to evaluate their relevance to determing whether can be recognised as a legal person. The paper argues that while AI does not satisfy certain all traditional requirements of legal personhood, some classical theories provide a persuasive jurisprudential basis for recognizing a limited and functional form of AI legal personality. Through this analysis, the study contributes to the ongoing discourse on adapting legal systems to the challenges posed by artificial intelligence in the digital age.
[1]Chinimilli Venkata Rama Padmaja et al., “The Rise of Artificial Intelligence: A Concise Review”, IAES International Journal of Artificial Intelligence (IJ-AI) 13(2) 2226, 2226 (2024).
[2]Tiing Leong Ang et al., “The Rise of Artificial Intelligence: Addressing the Impact of Large Language Models Such as ChatGPT on Scientific Publications”, Singapore Medical Journal 64(4) 219, 219 (2023).
[3]Salomon v. A. Salomon & Co. Ltd. [1897] AC 22 (HL).
[4]Manohar Ganesh Tambekar v. LakhmiramGovindram, (1888) ILR 12 Bom 247; Yogendra Nath Naskar v. Commissioner of Income Tax, AIR 1969 SC 1089.
[5]Shiromani Gurdwara Parbandhak Committee v. Som Nath Dass,(2000) 4 SCC 146.
[6]Te Awa Tupua (Whanganui River Claims Settlement) Act 2017 (NZ), s. 14;Centro de Estudios para la Justicia Social Tierra Digna y Otros v. Presidencia de la República y Otros, Sentencia T-622/16 (Constitutional Court of Colombia, Nov. 10, 2016).
[7]See generally Shawn Bayern, “The Implications of Modern Business-Entity Law for the Regulation of Autonomous Systems”, Stanford Technology Law Review 19(1) 93, 94–96 (2015); European Parliament, Report with Recommendations to the Commission on Civil Law Rules on Robotics, 2017/2103(INL) (Jan. 27, 2017), para. 59(f); UNESCO, Recommendation on the Ethics of Artificial Intelligence (adopted Nov. 23, 2021, UNESCO, 2022).
[8]Rowena Rodrigues, “Legal and Human Rights Issues of AI: Gaps, Challenges and Vulnerabilities”, Journal of Responsible Technology 4 100005, 1 (2020).
[9]See generally M. Mohit Prem Kumar, “Treating Artificial Intelligence as a Legal Person to Hold Intellectual Property under the Indian Legal Framework and Ethical Analysis”, Indian Journal of Integrated Research in Law 5(1) 1048, 1060–1061 (2025); Aditi Bharti & Gagandeep Kaur, “Legal Personhood of Artificial Intelligence: A Contemporary Perspective on Juristic & Electronic Personality”, Educational Administration: Theory and Practice 30(5) 10395, 10399 (2024).
[10]See generally Yash Choudhary, “Legal Personhood for AI: A Possible Key for Unlocking Human-AI Symbiosis?”,RGNUL Student Research Review (Dec. 28, 2022), available at: https://www.rsrr.in/post/legal-personhood-for-ai-a-possible-key-for-unlocking-human-ai-symbiosis(last visited June 26.2026);SimonChesterman,“Artificial Intelligence and the Limits of Legal Personality,” 69(4) International and Comparative Law Quarterly 819, 833–836 (2020).
[11]Rowena Rodrigues, supra note 8.