<![CDATA[The rapid integration of generative artificial intelligence into knowledge-sensitive domains has prompted extensive ethical debate, largely focused on issues of accuracy, bias, and governance. This article argues that such approaches overlook a more fundamental normative challenge: the transformation of epistemic responsibility and authority in contemporary practices of knowing. Drawing on epistemic ethics and virtue epistemology, the article contends that generative AI systems lack epistemic agency and therefore cannot bear responsibility for knowledge claims. Through normative conceptual analysis, the article identifies a structural "responsibility gap" that emerges when epistemic reliance is conflated with trust. Furthermore, this article provides a theoretical in-depth analysis of human-AI assemblages and proposes a practical implementation framework for higher education and journalism. The framework emphasizes accountable agency, institutional responsibility, and the cultivation of epistemic virtues as necessary conditions for integrating generative AI without undermining democratic epistemic practices.]]>
Copyrights © 2026
