본문 바로가기
  • Home

AI Ethical Alignment and AGI Ethical Alignment: Comparison and Research Directions

  • Journal of Software Assessment and Valuation
  • Abbr : JSAV
  • 2025, 21(4), pp.79~88
  • Publisher : Korea Software Assessment and Valuation Society
  • Research Area : Engineering > Computer Science
  • Received : November 7, 2025
  • Accepted : December 20, 2025
  • Published : December 26, 2025

Hojung Lim 1 Yoo, Seong joon 2 Jae Yoo Lee 2

1한국전자기술연구원
2세종대학교

Accredited

ABSTRACT

AI technology is driving structural changes across society, but the emergence of AGI (Artificial General Intelligence) reveals the limitations of existing extrinsic regulatory models, demanding embedded ethics. This paper defines AGI Ethical Alignment (AGI-EA) and analyzes technological-ethical structural differences through literature review and gap analysis. Centering on AGI-EA's four functional axes-(Value Embedding, Self-Judgment, Contextual Integration, and Moral Reasoning)-we propose a novel '6-Layer Adaptive Framework' (Value, Context, Social, Correction, Governance, Audit) to address AGI's self-evolution and emergence. This framework operates as a 'Multi-scale Self-Alignment Loop' through which AGI performs self-learning, self-correction, and self-auditing. We present this mechanism to secure ethical autonomy and a sustainable social adaptation structure. This study provides a theoretical blueprint for AGI-EA, serving as a foundation for technology-ethics integration strategies aligned with future international standards.

Citation status

* References for papers published after 2024 are currently being built.