Abstract
As ethical frameworks for artificial intelligence (AI) proliferate, gender equality emerges as a core ethical concern, yet its practical role in AI governance remains unclear. This study examines how gender equality is understood, prioritized, and enacted within AI ethics frameworks through qualitative interviews with 200 professionals across technology-related sectors on four continents, sponsored by UN Women. Utilizing thematic analysis, the research explores how ethical principles are cognitively processed and organizationally implemented by practitioners.
The findings reveal a consistent pattern where gender equality is formally acknowledged but often treated as a secondary issue, subordinate to technical priorities like data security, algorithmic transparency, and system performance. Interviewees frequently described gender equality as abstract, difficult to operationalize, or outside their primary responsibilities. As a result, engagement with gender-related ethical concerns was often symbolic, manifesting as compliance-oriented or box-ticking practices rather than substantive interventions.
Drawing on a Behavioral Public Policy perspective, the study illustrates how cognitive biases, professional norms, and institutional choice architectures shape ethical prioritization within AI governance. Ethical issues that are measurable, salient, and aligned with technical expertise are systematically favored, while socially embedded concerns like gender equality receive less attention. These dynamics help explain the gap between the formal inclusion of gender equality in AI ethics frameworks and its limited impact on decision-making and implementation.
By emphasizing behavioral mechanisms in ethical governance, this research contributes to discussions on responsible AI and public policy, advocating for behaviorally informed AI ethics frameworks that integrate gender equality as a concrete and enforceable component.