نشریه دولت و حقوق
سال پنجم شماره 1 (پیاپی 15، بهار 1403)

In response to the rapid development of artificial intelligence (AI), various legal systems have created different regulatory and supervisory interventions for this technology. While some countries prioritize consumer protection by imposing stringent regulations, others adopt a more permissive approach to promote innovation. However, comparing different models of AI regulation can help find the right balance between innovation and protecting consumers against the risks of AI. This paper, using a descriptive-analytical method and a comparative study, seeks to answer the question of what Iran's regulatory policy on artificial intelligence is. An examination of AI regulatory models in the European Union, the United States, and China indicates that each legal system must organize this matter according to its own objectives, policies, requirements, and characteristics of the legal system and technology. Iran's legal system suffers from serious structural and normative gaps in responding to this technology and other emerging technologies like AI. In 2021, the Supreme Leader of Iran declared the development of AI technology and its related systems as one of the country's macro-policies. However, the policy of developing artificial intelligence without addressing these structural and normative gaps not only potentially endangers the rights of citizens but also seriously challenges the development of AI-based businesses.

Today, new technologies are evolving rapidly, and in the meantime, artificial intelligence offers many potential benefits to mankind with its ever-increasing progress. Such technology requires the processing and analysis of significant amounts of data, as well as access to large volumes of big data. However, the ethical issues of artificial intelligence in the digital space must be identified and taken into account to prevent abuse. Therefore, ethical principles should be considered in the artificial intelligence-based system and guidelines should be provided for users who implement artificial intelligence technologies; which can be called digital ethics. Therefore, the main question of the paper is "What are the moral risks of artificial intelligence and is it possible to create liability insurance for artificial intelligence-based systems?". The research method of this paper is theoretical and based on a library data collection tool, and the data analysis method is descriptive-analytical. According to the findings of the research, the most important ethical challenges of artificial intelligence include ethical concerns related to the collection and unauthorized use of personal data, which will lead to the violation of privacy and human dignity. On the other hand, one of the concerns is non-compliance with the ethical principles of bias and data security. As a result, while trying to develop ethical standards with monitoring and accountability for artificial intelligence and digital space, digital liability insurance suggested in utilization of artificial intelligence in the digital space of liability and negligence in digital environments. Providers of digital technologies may require specialized malpractice insurance to address potential liability issues specific to these environments. Insurers must develop policies that reflect the unique risks associated with digital environments.

In today's world, artificial intelligence is increasingly gaining attention due to its ability to solve many real-world problems. This technology has penetrated various fields such as medicine, banking, and manufacturing. Despite the significant expansion of artificial intelligence in these fields, it does not have a prominent position in law yet. Artificial intelligence (AI) is increasingly driving advances in technology and business. It is widely used in various industries and the effects of artificial intelligence can be found in almost every aspect of creativity. The development of artificial intelligence is facilitated by large volumes of data and increasing available computing power. Legislators should make the necessary effort to identify it, especially since artificial intelligence has brought various aspects of the legal world under its influence, such as competition and business rights. In competition law, we try to make a fair balance in obtaining profit and economic benefit between different words, but this new technology has challenged the principles of competition law in practice. it should not be interpreted as a legal problem. We should obtain maximum utilization by creating a peaceful legal coexistence between these two institutions. This paper seeks to investigate the existing gaps by comparative study.

Artificial intelligence (AI) is rapidly transforming our world, and there is no exception even in the field of employment and labor rights. The pervasiveness of AI in tasks and activities previously performed by humans has raised concerns about the displacement of human labor and its implications for employment, labor rights, and social stability. AI has the potential to automate many tasks, which could lead to the loss of some jobs. However, AI can also create new jobs and enhance existing performance. The overall impact of AI on employment will depend on various factors, including the rate of automation, workforce skills, and government support policies. The emergence of AI in the workspace presents several legal and ethical challenges, including algorithmic discrimination, data privacy breaches, and accountability, that require addressing. This paper adopts a descriptive-analytical approach to provide legal solutions to address AI challenges in the field of work. These solutions include amending existing laws, enacting new regulations, raising awareness and education, and protecting workers' rights. The aim of this paper is to provide a comprehensive overview of the legal issues arising from AI in the context of labor rights and to offer practical solutions to ensure workers' rights in the AI era.

Artificial intelligence (AI), with astonishing progress, is exploiting abilities that were previously under human dominance. Regardless of technical complications, this enigma requires a comprehensive analysis of humanities elites to respond to emerging challenges. One of these challenges in the field of law is the AI liability. If damage is caused by AI, then who should be held liable? Noting that Iran's legal system currently does not have specific regulations on AI, following similar issues, it is argued that the user of AI is liable based on fault and the liability of the producer is vague. A comparative study of the European Union regulations shows that the liability of the user is based on fault, with a slight difference, while the liability of the producer is absolute. However, none of these legal systems have given a separate character to AI and do not consider it as more than one object. This paper considers the unique characteristics of AI and believes that both of the reviewed regulations are ineffective. Therefore, it is appropriate to define an independent personality for AI to prevent the infringement of the injured rights.

Artificial intelligence is built on technologies such as machine learning and deep learning neural networks, whose performance relies on a vast amount of high-quality data. Personal data is an indispensable resource for the development of artificial intelligence and the improvement of the efficiency of AI algorithms. Personal data is attributed to individuals and part of an individual's rights and obligations in the analog space can, depending on the case, have a close connection with its proper processing, which ensures the benefits for the data subject and prevents harm against them. Before the advent of artificial intelligence, the volume of personal data processing was relatively small and limited compared to AI; thus, the General Data Protection Regulations (2016) has mandated duties for each instance of personal data processing for the controller and has also determined rights for the data subject regarding each instance of personal data processing. However, considering the technical nature of artificial intelligence, the application of these rules to the algorithmic processing of personal data is fraught with difficulties. This article examines the challenges of the GDPR in protecting personal data under the processing of AI systems and analyzes legislative solutions for protecting personal data in the field of artificial intelligence. The major question of this article is, what is the legislative solution to overcome the challenge of personal data protection in algorithmic processes? This research is conducted using a descriptive-analytical method.

Artificial intelligence as an emerging technological field is rapidly developing and exploiting globally. The normative and regulation of this technological field has both a positive effect on its growth rate and in other side increases the level of its exploitation by reducing possible negative effects. Achieving these two goals, i.e., accelerating the speed of growth and simultaneous controlling side effects to increase the amount of use of artificial intelligence, through the lever of normative and regulation, requires a guiding macro view. Based on this necessity, the current research seeks to answer the question of how such a normalization process is achieved in the field of artificial intelligence. A descriptive-analytical method has been used. for this purpose. Therefore, in order to cover all aspects, the current research considers the broad meaning in relation to standardization; Therefore, artificial intelligence standardization is defined as a set of laws, rules, regulations and everything trying to somehow bind the field of the technology. Then, the concept of doctrine has been introduced as an upstream policy tool to coordinate and synergize multiple norms emanating from multiple sources. In the next step, it is discussed that the normative process is affected by the socio-political context and the stage of artificial intelligence development. Therefore, the guiding doctrine of artificial intelligence standardization during the evolution of this technology in the country is not fixed and it evolves depending on the conditions. In this regard, four stages have been introduced for the process of doctrinal evolution, including directive, descriptive, structural and positive.In this doctrinal evolution,the general approach changes from a recommendation paradigm to a norm prescription paradigm.In the first stage,guidance doctrine prevails,that is,guidance norms that are currently non-mandatory are presented.