The Need for Governance in the Field of Artificial Intelligence (AI)

Currently, Artificial Intelligence (AI) has been generating intense debates, either because there are divergent understandings of what the term is or because there are several speculations about its future. Regardless of the conceptual multiplicity, it is inevitable that this technology will be gradually incorporated into everyday life since it holds enormous potential to bring benefits to society in different areas, such as education, science, management, and decision-making (COECKELBERGH, 2020; DAFOE, 2018; MÄNTYMÄKI, et al. 2022).

However, governments and private companies struggle to deal with this new reality. It is because it is usual that new issues arise with the birth of a technology, among them ethical matters. In this sense, although AI develops actions with ethical repercussions, this technology has no moral responsibility for its own actions. If, on the one hand, AI benefits society, as in helping diagnose certain diseases – such as cancer and Alzheimer’s -, for example, on the other hand, the increasing use of this technology has brought with it a new range of ethical and moral problems that need to be considered (DAFOE, 2018; GASSER; ALMEIDA, 2017; TAEIGAGH, 2021).

The use of AI is not limited to individual spheres but encompasses and impacts social and political structures. Because of this, it is the role of governments and the Public Administration to move from a state of lack of policy guidance regarding the use of AI to one of readiness to deal with this new technology through specific management mechanisms. This will enable the Public Administration to be able to mitigate the risks of AI while making the most of its benefits (DAFOE, 2018; MÄNTYMÄKI, et al. 2022; TAEIGAGH; 2021).

One such management mechanism is governance. Gahnberg (2021) understands governance in AI as a set of intersubjectively recognized rules that define, limit, and shape expectations about the fundamental properties of an artificial agent. Meanwhile, Butcher & Beridze (2019) understand the concept as a multiplicity of instruments and solutions capable of influencing the development and applicability of AI. In turn, Mantymäki et al. (2022) conceptualize governance in the AI field as a system of rules, practices, processes, and tools used to ensure that the organization’s use of AI aligns with its own strategies, values, and goals; as well as fulfilling the legal requirements and AI ethical principles followed by it.

The study of governance in the field of AI allows us to list certain consensuses. First, the current literature on AI governance is underdeveloped and disorganized, with gaps to be filled and an impossibility of reaching a single definition, either because there are multiple definitions of what governance is, or even of what is meant by AI, or because of the complexity of the phenomenon (UZUN, et al. 2022). Second, the importance of governance in the field of AI is very high since it holds the ability to manage the risks brought by this technology and, at the same time, reduce the risks brought by these systems (MÄNTYMÄKI, et al. 2022). Third, the study of governance in AI is in its early stages. Governments perceive the need for a clear position before this reality, given the importance of the topic on their agendas, because it is directly related to the quality of life of future generations (DAFOE, 2018).


COECKELBERGH, Mark. AI ethics. MIT Press, 2020.

DAFOE, Allan. AI governance: a research agenda. Governance of AI Program, Future of Humanity Institute, University of Oxford: Oxford, UK, v. 1442, p. 1443, 2018.

GASSER, Urs; ALMEIDA, Virgilio AF. A layered model for AI governance. IEEE Internet Computing, v. 21, n. 6, p. 58-62, 2017.

GAHNBERG, Carl. What rules? Framing the governance of artificial agency. Policy and Society, v. 40, n. 2, p. 194-210, 2021.

TAEIHAGH, Araz. Governance of artificial intelligence. Policy and Society, v. 40, n. 2, p.137-157, 2021. UZUN, Mehmet Metin; YILDIZ, Mete; ÖNDER, Murat. Big Questions of AI in Public Administration and Policy. Siyasal: Journal of Political Sciences, v. 31, n. 2, p. 423-442. 2022

Back To Top