EUVIMEDSkopri EUVIMED
Open access · CC BY 4.0

From Black Boxes to Glass Boxes: Explainable AI for Trustworthy Deepfake Forensics

Hanwei Qian, Lingling Xia, Ruihao Ge, Yiming Fan, Qun Wang, Zhengjun Jing

Cryptography · 2025-09-26 · 9 (4) · 61

DOI: 10.3390/cryptography9040061 ↗

At a glance

Journal article in Cryptography, published online on 2025-09-26. Written by Hanwei Qian and 5 co-authors. Cited 10 times so far (Crossref); 53 references. The full text is freely available (open access, CC BY 4.0). Funded by: MOE Industry-University Collaborative Education Project, Philosophy and Social Sciences Research Project of Universities in Jiangsu Province, “Cyberspace Security” construction project of key disciplines in Jiangsu Province during the “14th Five-Year Plan”.

AstrattDettalji biblijografiċiCiteRelated studies

Abstract (original language)

As deepfake technology matures, its risks in spreading false information and threatening personal and societal security are escalating. Despite significant accuracy improvements in existing detection models, their inherent opacity limits their practical application in high-risk areas such as forensic investigations and news verification. To address this gap in trust, explainability has become a key research focus. This paper provides a systematic review of explainable deepfake detection methods, categorizing them into three main approaches: forensic analysis, which identifies physical or algorithmic manipulation traces; model-centric methods, which enhance transparency through post hoc explanations or pre-designed processes; and multimodal and natural language explanations, which translate results into human-understandable reports. The paper also examines evaluation frameworks, datasets, and current challenges, underscoring the necessity for trustworthy, reliable, and interpretable detection technologies in combating digital misinformation.

Funding

  • MOE Industry-University Collaborative Education Project
  • Philosophy and Social Sciences Research Project of Universities in Jiangsu Province
  • “Cyberspace Security” construction project of key disciplines in Jiangsu Province during the “14th Five-Year Plan”

Cite

APA

Qian, H., Xia, L., Ge, R., Fan, Y., Wang, Q., & Jing, Z. (2025). From Black Boxes to Glass Boxes: Explainable AI for Trustworthy Deepfake Forensics. Cryptography, 9(4), 61. https://doi.org/10.3390/cryptography9040061

Vancouver

Qian H, Xia L, Ge R, Fan Y, Wang Q, Jing Z. From Black Boxes to Glass Boxes: Explainable AI for Trustworthy Deepfake Forensics. Cryptography. 2025;9(4):61. doi:10.3390/cryptography9040061

BibTeX
@article{Qian2025,
  author = {Qian, Hanwei and Xia, Lingling and Ge, Ruihao and Fan, Yiming and Wang, Qun and Jing, Zhengjun},
  title = {From Black Boxes to Glass Boxes: Explainable AI for Trustworthy Deepfake Forensics},
  journal = {Cryptography},
  year = {2025},
  volume = {9},
  number = {4},
  pages = {61},
  doi = {10.3390/cryptography9040061}
}
Start for free →