Research on Generative AI Content Falsification Detection Algorithms and Enterprise Data Traceability Technologies

Jialin Han

doi:10.71222/nmxycp08

Authors

Jialin Han FESCO Adecco Human Resources Service Shanghai Co., Ltd., Shanghai, China Author

DOI:

https://doi.org/10.71222/nmxycp08

Keywords:

deepfake detection, multimodal forensics, consortium blockchain, data traceability, generative ai

Abstract

Generative large-scale models continue to evolve rapidly, significantly enhancing their capabilities in automated multimodal content generation. Highly realistic synthetic content is increasingly permeating various enterprise data flow scenarios, posing substantial threats to data credibility management and compliance governance systems. Existing deepfake detection algorithms generally suffer from poor generalization capabilities, struggling to identify novel forged samples generated by diverse model architectures and failing to meet authenticity verification requirements in complex operational contexts. Moreover, enterprises face dual security risks of unauthorized data alteration and core privacy breaches during cross-party data sharing and circulation. Addressing these industry challenges, this study develops a geometric-semantic decoupling dual-branched forgery detection model grounded in multimodal forensics theory and distributed trusted storage technology. By systematically eliminating semantic interference features, the proposed model accurately identifies unique forgery residual patterns inherent in generative content. Furthermore, by integrating a consortium blockchain architecture with threshold-based proxy re-encryption algorithms, the research establishes a privacy-preserving end-to-end data traceability system tailored for commercial applications, achieving integrated management of multimodal content authenticity verification, data circulation documentation, and ownership accountability. Experimental results demonstrate that the proposed model achieves an average AUC value of 94.3% across various AI-generated forgery samples, with traceability system latency below 12 ms per data record on-chain, effectively improving data processing efficiency while maintaining high identification accuracy. This research provides actionable theoretical foundations and practical technical solutions for safeguarding digital assets and ensuring compliance management of generative AI content in enterprise environments.

References

1. W. Xie, W. Li, and H. Zhang, "Electronic voting privacy protection scheme based on double signature in consortium blockchain," in International Conference on Artificial Intelligence Security and Privacy, Singapore, Dec. 2023, pp. 548–562, Springer Nature Singapore.

2. C. Peng, J. Li, D. Liu, N. Wang, R. Hu, and X. Gao, "Deepfake detection in the era of large models," Sci. Sin. Inf., vol. 56, no. 1, p. 1.

3. S. M. Qureshi, A. Saeed, S. H. Almotiri, F. Ahmad, and M. A. Al Ghamdi, "Deepfake forensics: a survey of digital forensic methods for multimodal deepfake identification on social media," PeerJ Comput. Sci., vol. 10, p. e2037, 2024.

4. J. T. Lou, S. A. Bhat, and N. F. Huang, "Blockchain-based privacy-preserving data-sharing framework using proxy re-encryption scheme and interplanetary file system," Peer-to-Peer Netw. Appl., vol. 16, no. 5, pp. 2415–2437, 2023.

5. D. Salvi, H. Liu, S. Mandelli, P. Bestagini, W. Zhou, W. Zhang, and S. Tubaro, "A robust approach to multimodal deepfake detection," J. Imaging, vol. 9, no. 6, p. 122, 2023.

6. S. Jia, R. Lyu, K. Zhao, Y. Chen, Z. Yan, Y. Ju, and S. Lyu, "Can chatgpt detect deepfakes? a study of using multimodal large language models for media forensics," in Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit., 2024, pp. 4324–4333.

7. S. Nailwal, S. Singhal, N. T. Singh, and A. Raza, "Deepfake detection: A multi-algorithmic and multi-modal approach for robust detection and analysis," in *2023 Int. Conf. Res. Methodol. Knowl. Manag., Artif. Intell. Telecommun. Eng. (RMKMATE)*, Nov. 2023, pp. 1–8, IEEE.

8. P. Liu, Q. Tao, and J. T. Zhou, "Evolving from single-modal to multi-modal facial deepfake detection: Progress and challenges," arXiv preprint arXiv:2406.06965, 2024.

9. M. Lomnitz, Z. Hampel-Arias, V. Sandesara, and S. Hu, "Multimodal approach for deepfake detection," in 2020 IEEE Appl. Imagery Pattern Recognit. Workshop (AIPR), Oct. 2020, pp. 1–9, IEEE.

10. D. Tan, Y. Yang, C. Niu, S. Li, D. Yang, and B. Tan, "A review of deep learning based multimodal forgery detection for video and audio," Discover Appl. Sci., vol. 7, no. 9, p. 987, 2025.

11. M. A. Raza and K. M. Malik, "Multimodaltrace: Deepfake detection using audiovisual representation learning," in Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit., 2023, pp. 993–1000.

12. B. Liu, B. Liu, M. Ding, and T. Zhu, "ForgeFinder: Perceptive Multimodal Deepfake Detection via Multi-grained Forgery Localization," ACM Trans. Multimedia Comput. Commun. Appl., vol. 22, no. 1, pp. 1–24, 2026.