From Responsible Leadership to Responsible AI Governance: A Leadership-Based Framework for AI Governance in Organizations

Artificial intelligence (AI) is reshaping organizational decision-making, managerial roles, and accountability, yet responsible AI governance remains dominated by principles, technical controls, and regulatory compliance. This conceptual paper develops a human leadership-based framework for translating responsible AI principles into organizational governance. In the framework, the responsible leader is an accountable human organizational actor—such as a board member, senior executive, business leader, or designated decision owner—while AI is treated as a governed technological system or operational agent rather than the accountable leader. Human responsible leadership influences responsible AI governance (RAIG) capability through six mediating organizational processes: stakeholder orientation, ethical sensemaking, accountability architecture, meaningful human oversight, inclusive participation, and long-term value orientation. RAIG capability is defined as the higher-order organizational capacity to institutionalize and adapt lifecycle governance arrangements; it enables observable governance practices, which are conceptually separated from downstream responsible AI outcomes. AI operational autonomy is proposed as a focal moderator: the leadership-to-RAIG pathway should become stronger when AI can initiate or execute consequential actions with limited contemporaneous human approval, particularly where actions are difficult to reverse, trace, or interrupt. The framework offers testable propositions and a research agenda, but it is conceptual and requires empirical validation; it is not a technical safety guarantee.