The advancement of artificial intelligence (AI) has raised questions about its potential utilization in various roles, including as a member of the board of directors in a Limited Liability Company. In Indonesia, this phenomenon was highlighted when PT Suryadhamma Investama claimed to have appointed an AI named Ardi as its Director. This study aims to examine the legal feasibility of AI serving as a corporate director within the Indonesian legal framework. Employing a doctrinal research method, this study relies on secondary data obtained from literature studies. The analysis is conducted using Article 1367 paragraph (1) of the Civil Code to explore the legal status of AI and Article 93 paragraph (1) of the Company Law to examine the normative requirements for the position of a Director. The findings indicate that, under Article 1367 paragraph (1) of the Civil Code, AI can be analogized as a legal object categorized as goods under the supervision of its creators, sponsors, and owners. Furthermore, based on Article 93 paragraph (1) of the Company Law, the role of a Director is normatively restricted to natural persons who meet specific eligibility criteria, thereby excluding AI from such appointments. These findings have implications for corporate governance and the legal responsibilities associated with AI utilization, emphasizing the need for regulatory clarity in addressing AI's role within organizational structures.

Keywords: Company Director, Artificial Intelligence, Legal Subject.

INTRODUCTION

The presence of AI has been predicted by experts, including Stephen Hawking, to have an effect on the social order of society. One important context to discuss relates to the position of AI as a legal subject. Because if a machine can think and act like a human, shouldn't AI also be recognized as an entity with rights and obligations, namely as a legal subject? (Sultonova et al., 2023).

In some countries, AI has been recognized as a legal subject with certain capacities (Amboro & Komarhana, 2021). For example, Saudi Arabia granted an AI-powered robot named Sofia Saudi Arabian citizenship in 2017 (Santoni de Sio & Mecacci, 2021). Japan granted a residence permit to an AI-powered robot named Shibuya Mirai in 2017. Even the Russian Parliament has drafted the Grishin Law in 2015, which amends the provisions of the Civil Code of the Russian Federation (Kusumawardani, 2019). The amendments impose legal liability on robot developers, operators, or manufacturers, and regulate the issue of representation of AI-powered robots in court. These conditions indicate that there is a fairly rapid development of technology, so the legal field must certainly keep up with developments in order to regulate and ensure that its implementation does not harm the interests of society. This includes Indonesia, which will one day be affected by these developments.

Until now, the discussion about AI in Indonesia has not yet reached an agreement in terms of its position as a bearer of rights and obligations in the eyes of the law (Mahardika, 2021). Indonesia does not yet have laws that specifically regulate these smart machines. However, when viewed from its characteristics, AI that is capable of processing information automatically makes it equal to Electronic Agents in Indonesian legislation. Based on Article 1 paragraph 8 of Law Number 19 of 2016 concerning Amendments to Law Number 11 of 2008 concerning Electronic Information and Transactions (here in after abbreviated as ITE Law), an Electronic Agent is a device of an Electronic System that is made to perform an action on certain Electronic Information automatically organized by a Person. The word "automatically" in the definition is then used as a basis for constructing AI as part of an Electronic Agent.

The use of AI is currently rampant in order to maximize the potential of existing technology. The utilization of AI that will be focused on in this study is the possibility of AI to be used as one of the organs of a Limited Liability Company, namely as a member of the board of directors (Amboro & Komarhana, 2021). Among the phenomena of utilizing AI to assist management activities in Limited Liability Companies, there is an algorithm engine called VITAL (Validating Investment Tool For Advancing Life Sciences) which was appointed as a member of the board of directors of the Hong Kong-based Deep Knowledge Ventures company. Another practical example in 2016, Tieto, a European company, appointed an automated AI machine named Alicia T as one of the members with full voting rights in the leadership team. In Indonesia, the phenomenon of utilizing AI as an organ of the company also occurred, where PT Suryadhamma Investama claimed to have appointed artificial intelligence technology named Ardi as its Director. So based on Indonesian law, is AI a legal subject, so it is possible to be appointed as a director of the company?

This study offers a novel perspective by critically examining AI's position within Indonesian corporate law. Unlike previous studies that focus primarily on AI's ethical or technological aspects, this research delves into the legal feasibility of appointing AI as a corporate director under Indonesian law. By analyzing the intersection of AI capabilities, legal constructs such as Electronic Agents, and corporate governance frameworks, this study provides a unique contribution to understanding the evolving role of AI in organizational structures.

Based on the background above, the purpose of this research is to analyze the legal feasibility of AI serving as a director under Indonesian law. The findings of this research are expected to benefit stakeholders by clarifying the legal status of AI and providing guidance for policymakers and corporate entities in adapting to technological advancements responsibly.

RESEARCH METHOD

This study employs a doctrinal research method to analyze the legal feasibility of artificial intelligence (AI) serving as a corporate director within the framework of Indonesian law. Using secondary data sourced from legal statutes, academic literature, and international case studies, the analysis focuses on two key aspects: the classification of AI as a legal object under Article 1367 paragraph (1) of the Civil Code and its eligibility as a director based on Article 93 paragraph (1) of the Company Law, which requires directors to be natural persons with legal capacity. By applying qualitative legal analysis, this study systematically interprets these legal provisions, compares them with global practices, and evaluates the implications for corporate governance. The findings are expected to bridge gaps in the existing literature, offering guidance to policymakers and businesses in navigating the integration of AI into organizational structures while ensuring compliance with legal standards.

RESULT AND DISCUSSION

The Position of Artificial Intelligence as a Civil Law Subject

AI is an artificial intelligence that is said to be one of the most important inventions of the 21st century (Hasibuan et al., n.d.). AI is a technology that allows computers or machines through software to mimic human intelligence, so that they can act and think rationally like humans and have the ability to solve problems. AI is assumed to have a role as a dynamic entity that can think, understand, and adapt from the data presented (Winston & Holmes, 2018). The processed data is then recorded and stored so that it becomes knowledge for AI in making decisions later (Simbolon, 2023).

When viewed from a historical perspective, AI was officially created for the first time in 1956 by Marvin Minsky and John McCarthy at the Darmouth Summer Research Project on Artificial Intelligence (DSRPAI) workshop held at the University of Darmouth, New Hampshire (Haenlein & Kaplan, 2019). DSRPAI aims to bring together researchers from various fields of science to make the latest discovery, which is to create a machine that can simulate human intelligence. This workshop was followed by other inventions in the field of AI, such as the invention of ELIZA in 1964-1966, a computer program created by Joseph Weizenbaum that produced a programming language that could communicate with humans. Another invention was the General Problem Solver Program created by Cliff Shaw and Allen Newel, which can solve simple problems automatically. Research into AI continues to this day, resulting in increasingly sophisticated AI.

When viewed from a legal perspective, opinions regarding AI as a legal subject are currently still divided into at least 2 (two) contrasting opinions. Some experts consider that AI is an object of law, while on the other hand some experts believe that AI is a subject of modern law (Sultonova et al., 2023). The arguments of experts, one of which is Prof. Nils John Nilsson from Stanford University (Nilsson, 2009) in favor of the concept of AI as a legal object or tool is based on: legal rules and norms are made for humans and are based on human needs; and legal norms always consider the characteristics of humans who can feel and sensitivity to things that happen (Sultonova et al., 2023).

The concept of AI as a legal object or tool under the responsibility of a legal subject has its positive and negative sides. On the positive side, AI as a legal object can make it easier to identify liability if there is damage caused by AI. This means that the organizer of the AI is responsible for all its actions. But on the other hand, it can ignore the growing capabilities of AI, such as being able to make decisions independently, assist research, and even produce intellectual products.

Some other experts agree that AI should be treated as a legal subject, and carry all the rights and responsibilities of a human or legal entity. One such opinion is expressed by Kakoudaki, who defines AI's legal personality as similar to the legal capacity of workers in ancient times, with very limited legal status and no rights (Oleksiewicz & Civelek, 2019).. Another pro opinion was also expressed by Prof. Saldi Isra, who is one of the Judges of the Constitutional Court of the Republic of Indonesia. Prof. Saldi argues that AIs should be considered as legal subjects with rights and responsibilities like humans, based on the principle that AIs can act independently and have a major impact on society and the environment (KlikLegal.com, 2023).

When referring to Article 1367 paragraph (1) of the Civil Code, there is a relationship between workers, employers, and goods under their supervision (Wetboek et al., 2014), which reads as follows:

"(1) A person is liable not only for damages caused by his own acts, but also for damages caused by the acts of those for whom he is responsible or by goods under his supervision."

Based on this article, AI can be analogized to goods under the supervision of an employer. The use of AI in everyday life is intended to assist human work. AI is born with previously embedded knowledge in the form of memory or data, and is designed to perform certain specific tasks called algorithms (Sultonova et al., 2023). AI seems to be able to make decisions and think logically like humans, but in reality the decisions given by AI are the result of collaboration from the data stored in it. These data are entered by the creator or manager of AI, so the knowledge contained in AI is limited only to the extent that has been entered into it.

With this line of thinking, it can be concluded that the creator or manager of AI is responsible for all decisions made by AI. Thus, AI is actually a legal object whose use is part of the responsibility of the legal subject, namely the person who controls it.

This opinion is in line with that expressed by Prof. Ahmad M. Ramli, who conducted a review of the AI law applicable in the European Union, namely the European Union Artificial Intelligence Act 2024 (hereinafter referred to as "AIA 2024") (Ramli, 2024). Based on Article 14 paragraph (1), Article 50 paragraph (1), and Article 57 paragraph (12) of AIA 2024, the European Union explicitly stipulates that AI is a legal object whose use must be supervised by humans, so that humans have full control over the decisions made by AI. The liability for any harm caused by AI is the responsibility of the human entity, such as the developer, operator, or party controlling the AI, and not the responsibility of the AI itself.

The idea of giving AIs the same legal status as humans is considered to destroy the existing legal system and foundations of the state (Santoni de Sio & Mecacci, 2021). An example is the AI-powered robot created by Hanson Robotics named Sofia, who was granted citizenship of Saudi Arabia in 2017. The decision caused conflict and criticism from various parties. One of the conflicts that arose was that Saudi Arabian law requires its female citizens to wear the hijab, and can travel far if they have been permitted by their husbands or relatives. Of course, this legal norm cannot be applied to robots. Criticism also arises because the granting of citizenship to Sofia was done in a very short time without any problems, while on the other hand there are still many immigrants who live or work there who cannot obtain Saudi citizenship and their rights are very limited.

Humans are highly complex beings with a mind, soul, subconscious, reason, desires, feelings, and emotions. Whereas AI, despite the lure of being a human clone, is merely an imitation machine, which lacks the real "life" aspect that exists in the human psyche. AI does not have free will, due to its limited capabilities and is highly dependent on the person who controls it to "input" knowledge into its system. Based on these considerations, AI is a legal object, and its subjects are parties who have a legal relationship with the object, including the creator, sponsor and owner of the AI system (Sultonova et al., 2023).

Artificial Intelligence as a Director of the Company in Review of the Applicable Law in Indonesia

The Board of Directors is an organ of the company that is authorized to represent the company, enter into agreements or contracts, and so on (Samadani, 2013). Based on the theory of corporate law, there are 2 (two) main duties of a director of the Company, namely duty of care and duty of loyalty. These duties include preparing and analyzing information relating to the scope of the company's business, business competition, market conditions, and monitoring the company's economic and financial prospects. In analyzing this information, it must use the business judgment rule, where in making a decision must be based on specific information and good analysis (Kamalnath, 2019). AI as software that is able to analyze and provide concrete decisions based on the available data, certainly at first glance, seems to be able to meet these criteria. However, if we look at the construction of the Company Law, especially the articles relating to company directors, it can be indicated that AI normatively cannot be used as a director in a limited liability company.

Formally, the regulation of the Company's directors is based on the Limited Liability Company Law and the Company's Articles of Association, while materially it is based on regulations related to the company's business activities. (Utamy & Wahjuni, 2020). Based on Article 93 paragraph (1) of the Company Law, it is stated that those who can be appointed as members of the board of directors are individuals who are capable of performing legal acts, and within 5 (five) years before their appointment have never been:

Declared bankrupt;
Being a member of the board of directors or a member of the board of commissioners who is found guilty of causing a company to be declared bankrupt; or
Convicted of a criminal offense that harms state finances and/or is related to the financial sector.

If we follow the construction of Article 93 paragraph (1) of the Company Law, then there are 3 (three) main requirements in order to be appointed as a member of the board of directors of a PT, namely:

Natural person

In law, two legal subjects are known, namely a private person (naturlijk person) and a legal entity (recht person). A personal person as a legal subject is born naturally. Legal authority is given to each person, so according to L.J. Van Apeldoorn everyone is a legal subject. In other words, only a naturlijk person or a private person can be appointed as a director.

Capable of performing legal acts

Article 1329 of the Civil Code states that "every person is capable of making obligations, unless he is determined to be incapable by law". From this article, it can be understood that although everyone is a supporter of rights and obligations as a legal subject, not everyone is capable of performing legal acts (legal capacity).

As for those who are not capable of making an agreement based on Article 1330 of the Civil Code, they are minors, persons placed under guardianship, and women who have married in cases determined by the Law are prohibited from making certain agreements. Adults according to Article 330 of the Civil Code are those who have reached the age of 21 (twenty-one) years or have been married before. However, the rules in Article 330 of the Civil Code have been updated with the issuance of SEMA Number 3 of 1963 in conjunction with Article 31 of Law Number 1 of 1974 concerning Marriage, stating that women who are bound in marriage are capable of carrying out their own legal actions.

The last circumstance in which a person is not legally competent is those who are under guardianship, such as people who are classified as insane, dumb, and blind. In addition, people who are classified as extravagant may be placed under guardianship based on a court decision.

Never been declared bankrupt, never been a member of the board of directors or a member of the board of commissioners who was found guilty of causing the company to be declared bankrupt, and never been convicted of a criminal offense that harms state finances / related to the financial sector.

Based on this legal construction, normatively, the position of Director can only be occupied by humans, namely natural persons who have fulfilled all the requirements in Article 93 paragraph (1) of the Company Law. AI will indeed not be affected by conflicts of interest in making decisions (unless the knowledge put into it is biased from the start) (Kamalnath, 2019). Since AI's decision making is based on the available data and not based on feelings, AI does not have to worry about making decisions that can harm those closest to it, and only focuses on decisions that can benefit the company. However, AI does not have the instincts and feelings that people in the business world have. In making a decision, the director must consider the interests of the company as well as the shareholders (Sardjono et al., 2014). However, these things cannot be obtained only from written data, but also based on conscience by paying attention to the surrounding environment.

Allowing AI to make full decisions without considering shareholders, employees, or even society in general is feared to disrupt corporate governance, and can have negative impacts, especially for the company itself (Kamalnath, 2019). Without good corporate governance, the transparency and accountability of the company will be jeopardized (Day, 2024). Without good corporate governance, the transparency and accountability of the company will be jeopardized this problem can grow and it is feared that it can harm the company and its shareholders in the long run.

Indeed, AI still needs humans to enter knowledge in the form of data into its memory, so that from this knowledge AI can think and analyze the problems that occur. So it is not right if AI is used as a legal subject, namely as a director who replaces humans, because AI still needs humans in its operation. It would be more appropriate if AI is used as a tool or legal object that can support the activities of the director in carrying out company management activities, not being a direct part of the company's management.

If analogous to an autonomous car or driverless car, which still needs a human to control the vehicle to prevent accidents. Then the director can be likened to the driver of the autonomous car. The director has full control and can override decisions made by the AI if the decision is believed to be wrong, to prevent losses.

One of the loss scenarios that can occur is when the AI system is hacked, resulting in confidential company data being leaked. This is evidenced by the increasing number of cybercrime attacks in Indonesia with the most significant spike occurring in 2021, which amounted to 1,400,000,000 (one billion four hundred million) attempts, which experienced a fourfold increase compared to the previous year (Susila & Salim, 2020). So it is important for AI managers to prioritize not only the functional and efficient use of AI to help company performance, but also the security of company data and the security of the AI system used.

In addition to paying attention to the security of the AI system used, the utilization of AI must also pay attention to the algorithm for the data entered into the AI system (Emilio Ferrara, 2023). AI can be compared to a child who learns from its surroundings. AI will make decisions based on the data entered into it, so it can be imagined if the data entered is data that leads only to one particular party. This can lead to the phenomenon of machine learning bias or algorithm bias, namely the decisions produced by AI become biased or not in accordance with the actual situation (E Ferrara, 2023). Algorithm bias can occur due to the influence of several factors, such as the data entered into the AI system, algorithm design, and human interpretation.

In the context of utilizing AI as a tool that can support the activities of directors in the company, algorithm bias is feared to be detrimental to the company, because it can reduce the accuracy of AI thinking and reduce its potential in providing advice and input to the company's directors. Therefore, it is important to periodically check for AI bias, to ensure that AI can be used as an objective and reliable tool to support the board of directors in making decisions that are fair and in accordance with the company's objectives.

Another way to avoid AI bias is to create rules that require AI creators to train and check AI algorithms, which can be combined as a guarantee with consumer safety testing protocols commonly used in physical products (Haenlein & Kaplan, 2019). Another issue that is important to regulate is creating a code of ethics for AI creators, such as taking an oath in the profession of doctors, lawyers, and even notaries. In addition to AI creators, it is also important to create regulations for AI users, so that their use does not violate the rules and harm others. This regulation can be outlined in legislation that specifically regulates the rights and obligations of AI creators and users, considering that until now there has been no specific regulation on AI in Indonesia.

CONCLUSION

This research concludes that, based on Article 1367 paragraph (1) of the Civil Code, AI can only be classified as a legal object under the supervision of its creator, sponsor or owner. AI only operates based on pre-programmed algorithms and input data, lacking free will and independence. Although AI demonstrates significant capabilities in logical decision-making and performing certain tasks, it remains completely dependent on human intervention for knowledge input and operational control. Therefore, AI does not meet the criteria to be considered a legal subject, including occupying the role of a director. Based on the legal construction of Article 93 paragraph (1) of the Company Law, the position of director is only reserved for natural persons who meet certain criteria. Directors are required to make decisions that balance the interests of the company and its shareholders, which involves not only logical reasoning but also conscience, emotional intelligence, and an understanding of the surrounding environment-qualities that AI lacks.

Future research contributions emphasize the importance of using AI as a support tool, not as a replacement for human directors. AI can assist directors in data analysis, operational efficiency, and unbiased decision-making, thereby improving management activities. However, regulatory frameworks should address the ethical, legal, and operational implications of AI in corporate governance. Future research could focus on developing guidelines for integrating AI into management structures, ensuring its use complements human decision-makers while maintaining accountability and ethical standards. Such insights can inform businesses and policymakers about utilizing AI responsibly in modern corporate environments.

REFERENCES

Amboro, F. L. Y. P., & Komarhana, K. (2021). Prospek Kecerdasan Buatan Sebagai Subjek Hukum Perdata Di Indonesia [Prospects of Artificial Intelligence As a Subject of Civil Law in Indonesia]. Law Review, 21(2), 145–172.

Day, M. (2024). Why is corporate governance so important? Www.Diligent.Com. https://www.diligent.com/resources/blog/why-corporate-governance-important-investors?_x_tr_sl=en&_x_tr_tl=id &_

Ferrara, E. (2023). Fairness and Bias in Artificial Intelligence: A Brief Survey of Sources, Impacts, and Mitigation Strategies. Sci. 2024; 6: 3.

Ferrara, Emilio. (2023). Fairness and bias in artificial intelligence: A brief survey of sources, impacts, and mitigation strategies. Sci, 6(1), 3.

Haenlein, M., & Kaplan, A. (2019). A brief history of artificial intelligence: On the past, present, and future of artificial intelligence. California Management Review, 61(4), 5–14.

Hasibuan, R. H., Rawung, A. J., Paranduk, D. M. D., & Wowiling, F. J. (n.d.). rtificial Intelligence In The Auspices Of Law: A Diverge Perspective. Mimbar Hukum, 36(1), 111–140.

Kamalnath, A. (2019). The Perennial Quest for Board Independence: Artificial Intelligence to the Rescue? Alb. L. Rev., 83, 43.

KlikLegal.com. (2023). Kedudukan Hukum Artificial Intelligence: Tantangan dan Perdebatannya. KlikLegal.Com. https://kliklegal.com/kedudukan-hukum-artificial-intelligence-tantangan-dan-perdebatannya/#:~:text=Pertama%2C AI dapat dianggap sebagai,yang dilakukan oleh AI tersebut

Kusumawardani, Q. D. (2019). Hukum Progresif Dan Perkembangan Teknologi Kecerdasan Buatan. Veritas et Justitia, 5(1), 166–190.

Mahardika, Z. P. (2021). Pengaturan Hukum Artifical Intelligence Indonesia Saat Ini. Www.Hukumonline.Com. https://www.hukumonline.com/berita/a/pengaturan-hukum-artifical-intelligence-indonesia-saat-ini-lt608b740fb22b7

Nilsson, N. J. (2009). The quest for artificial intelligence. Cambridge University Press.

Oleksiewicz, I., & Civelek, M. E. (2019). From Artificial Intelligence To Artificial Consciousness: Possible Legal Bases For The Human-Robot Relatıonships In The Future.

Ramli, A. M. (2024). Artificial Intelligence (AI): Subjek Hukum atau Objek Hukum? Www.Hukumonline.Com. https://www.hukumonline.com/klinik/a/iartificial-intelligence-i-ai--subjek-hukum-atau-objek-hukum-lt670eca701fc74/

Samadani, A. (2013). Dasar-Dasar Hukum Bisnis. Mitra Wacana Media, Jakarta.

Santoni de Sio, F., & Mecacci, G. (2021). Four responsibility gaps with artificial intelligence: Why they matter and how to address them. Philosophy & Technology, 34(4), 1057–1084.

Sardjono, A., Dewi, Y. K., Irawaty, R., & Pangaribuan, T. (2014). Pengantar Hukum Dagang. Jakarta: Raja Grafindo Persada.

Simbolon, Y. (2023). Pertanggungjawaban Perdata Terhadap Artificial Intelligence Yang Menimbulkan Kerugian Menurut Hukum Di Indonesia. Veritas et Justitia, 9(1), 246–273.

Sultonova, L., Vasyukov, V., & Kirillova, E. (2023). Concepts of legal personality of artificial intelligence. Lex Humana (ISSN 2175-0947), 15(3), 283–295.

Susila, M. E., & Salim, A. A. (2020). Cyber Espionage Policy and Regulation: A Comparative Analysis of Indonesia and Germany. Computers & Electrical Engineering, 81, 1.

Utamy, P., & Wahjuni, S. (2020). Pertanggungjawaban Direksi Perseroan Terbatas Dan Notaris Terhadap Surat Kuasa Direksi Tentang Pembangunan Infrastruktur Pemerintah. Jurnal Bina Mulia Hukum, 4(2), 196–208.

Wetboek, B., Subekti, R., & Tjitrosudibio, R. (2014). Kitab Undang-Undang Hukum Perdata. (No Title).

Winston, P. H., & Holmes, D. (2018). The Genesis Enterprise: Taking artificial intelligence to another level via a computational account of human story understanding.

Sherin Dinda Muthia, Tjhong Sendrawan (2025)

First publication right:

Asian Journal of Engineering, Social and Health (AJESH)

This article is licensed under:

	Page 1	Asian Journal of Engineering, Social and Health
	Page 1	Volume 3, No. 12 December 2024