Human-centricity in AI governance: A systemic approach

Abstract

Human-centricity is considered a central aspect in the development and governance of artificial intelligence (AI). Various strategies and guidelines highlight the concept as a key goal. However, we argue that current uses of Human-Centered AI (HCAI) in policy documents and AI strategies risk downplaying promises of creating desirable, emancipatory technology that promotes human wellbeing and the common good. Firstly, HCAI, as it appears in policy discourses, is the result of aiming to adapt the concept of human-centered design (HCD) to the public governance context of AI but without proper reflection on how it should be reformed to suit the new task environment. Second, the concept is mainly used in reference to realizing human and fundamental rights, which are necessary, but not sufficient for technological emancipation. Third, the concept is used ambiguously in policy and strategy discourses, making it unclear how it should be operationalized in governance practices. This article explores means and approaches for using the HCAI approach for technological emancipation in the context of public AI governance. We propose that the potential for emancipatory technology development rests on expanding the traditional user-centered view of technology design to involve community- and society-centered perspectives in public governance. Developing public AI governance in this way relies on enabling inclusive governance modalities that enhance the social sustainability of AI deployment. We discuss mutual trust, transparency, communication, and civic tech as key prerequisites for socially sustainable and human-centered public AI governance. Finally, the article introduces a systemic approach to ethically and socially sustainable, human-centered AI development and deployment.

Publics’ views on ethical challenges of artificial intelligence: a scoping review

“…In the body of literature included in the scoping review, human agency and oversight emerged as the predominant ethical dimension under investigation. This finding underscores the pervasive significance attributed to human centricity, which is progressively integrated into public discourses concerning AI, innovation initiatives, and market-driven endeavours [15,112]. In our perspective, the importance given to human-centric AI is emblematic of the "techno-regulatory imaginary" suggested by Rommetveit and van Dijk [35] in their study about privacy engineering applied in the European Union's General Data Protection Regulation.…”

“…These efforts were publicly announced as aiming to create a common understanding of ethical AI development and foster responsible practices that address societal concerns while maximizing AI's potential benefits [13,14]. The concept of human-centric AI has emerged as a key principle in many of these regulatory initiatives, with the purposes of ensuring that human values are incorporated into the design of algorithms, that humans do not lose control over automated systems, and that AI is used in the service of humanity and the common good to improve human welfare and human rights [15]. Using the same rationale, the opacity and rapid diffusion of AI have prompted debate about how such technologies ought to be governed and which actors and values should be involved in shaping governance regimes [1,2].…”

AI for Identity and Access Management (IAM) in the Cloud: Exploring the Potential of Artificial Intelligence to Improve User Authentication, Authorization, and Access Control within Cloud-Based Systems

“…The emphasis on transparency in the system for increasing trust and acceptance is particularly noteworthy. This echoes the sentiments in the cybersecurity community about the importance of transparency in fostering trust among users, a point highlighted by Samuel et al [47]. It suggests that transparent systems, which make their operations and data handling clear to users, can significantly improve user trust and acceptance, an aspect that is increasingly becoming vital in the age of data privacy concerns.…”

“…Users who are more concerned about privacy tend to perceive greater vulnerability regarding information security threats, and this influences their behavioral intentions toward adopting new authentication technologies like BBCA [10,46]. Trust in AI-enabled systems, extending beyond technical capabilities, is pivotal for adoption and requires a human-centric approach, aligning AI systems with human values and ethical principles [47,48]. User trust is influenced by a combination of socio-ethical considerations, technical features, and user characteristics, with the latter emerging as a dominant factor; it's essential for AI systems to be transparent, fair, and ethically designed to foster user trust [10,49].…”

Adopting and expanding ethical principles for generative artificial intelligence from military to healthcare

“…Governability is the ability of a system to integrate processes and tools which promote and maintain its capability and ensure meaningful human control 95 . Standards for the governability of AI systems, as established by the DOD and NATO, emphasize the importance of ensuring that while AI systems fulfill their intended functions, humans must retain the ability to identify and prevent unintended consequences.…”