EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Authenticity Crisis, Personality Replication Risks, and Labeling Governance in Generative-AI Deep Synthesis: Toward a Traceable and Rights-Sensitive Framework

Zhiyong Song

Journal of Psychology & Education · 30.06.2026 · 1 (5) · 32

DOI: 10.66581/tbyn7997 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Psychology & Education, online erschienen am 30.06.2026. Verfasst von Zhiyong Song. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Generative artificial intelligence has transformed deep synthesis from a specialized technique of media manipulation into an ordinary infrastructure for producing text, images, audio, video, virtual scenes, digital humans, and increasingly personalized AI agents. The ethical problem is therefore no longer limited to whether a specific image or video is fake. A deeper challenge concerns the technological blurring of three boundaries: the boundary between authentic and synthetic content, the boundary between a real person and a digital replica, and the boundary between human agency and automated proxy action. This article argues that the governance of AI-generated synthetic content should not be reduced to content labeling alone. Labeling is necessary because it helps users identify the artificial origin of media, but it is insufficient when synthetic content imitates a particular person’s voice, likeness, expressive style, behavioral patterns, or decision-making tendencies. In such cases, the ethical issue is not merely deception but the unauthorized occupation of a person’s communicative position. The article develops a three-level analysis of ethical risk: epistemic risks to public truth and social trust, personality-rights risks arising from digital replication, and responsibility risks generated by persona-based AI agents. Drawing on current governance instruments in China, the European Union, the United States, and international AI ethics frameworks, the article proposes a five-dimensional governance model: identifiability, provenance traceability, identity authorization, revocability, and accountability. The central claim is that deep synthesis governance must move from a narrow model of “AI-content disclosure” toward a broader framework of “synthetic personhood governance,” in which content origin, identity authorization, agentic behavior, and responsibility allocation are treated as interconnected problems.

Zitieren

APA

Song, Z. (2026). Authenticity Crisis, Personality Replication Risks, and Labeling Governance in Generative-AI Deep Synthesis: Toward a Traceable and Rights-Sensitive Framework. Journal of Psychology & Education, 1(5), 32. https://doi.org/10.66581/tbyn7997

Vancouver

Song Z. Authenticity Crisis, Personality Replication Risks, and Labeling Governance in Generative-AI Deep Synthesis: Toward a Traceable and Rights-Sensitive Framework. Journal of Psychology & Education. 2026;1(5):32. doi:10.66581/tbyn7997

BibTeX
@article{Song2026,
  author = {Song, Zhiyong},
  title = {Authenticity Crisis, Personality Replication Risks, and Labeling Governance in Generative-AI Deep Synthesis: Toward a Traceable and Rights-Sensitive Framework},
  journal = {Journal of Psychology & Education},
  year = {2026},
  volume = {1},
  number = {5},
  pages = {32},
  doi = {10.66581/tbyn7997}
}
Kostenlos starten →