Discovery of Deepfakes in Art
DOI:
https://doi.org/10.55630/dipp.2025.15.5Keywords:
Deepfakes, Generative Models, Art Forensics, GANs, Transformers, Multimedia AIAbstract
The proliferation of deepfakes—AI-generated or manipulated media—has transformed the landscape of contemporary art. Deep generative models, including GANs, VAEs, diffusion models, and Transformers, have enabled artists to explore new creative realms while simultaneously raising critical questions around authenticity, ethics, and detection. This paper presents a comprehensive analysis of deepfake technologies across five key media modalities: image, video, text, speech, and music. We examine the architectures that enable content creation, and the state-of-the-art techniques used for detection. Further, we evaluate detection accuracy, robustness, and practical implementation, incorporating diagrams, comparative tables, and performance formulas. This work aims to provide a balanced perspective on the opportunities and challenges posed by synthetic media in the artistic domain.References
Alrashoud, M. (2025). Ethical boundaries of AI-generated content in digital culture. Journal of Emerging Media and Society, 14 (2), 133–150.
Babael, F., Amini, S., & Chen, Y. (2025). Multimodal deepfake detection in crossmedia environments. In Proceedings of the International Conference on Multimedia Security (pp. 45–59).
Chakraborty, A., Nair, V., Singh, R., & Joshi, P. (2024). Hybrid attention GAN for consistent face swapping in artistic performances. In Proceedings of the International Conference on Creative AI (pp. 22–34).
Dolhansky, B., Bitton, J., Pflaum, B., Lu, J., Howes, R., Wang, M., & Ferrer, C. C. (2020). The DeepFake Detection Challenge (DFDC) dataset. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), 2020. https://ieeexplore.ieee.org/document/9156360
Gao, Y., Lin, Q., & Wu, T. (2024). StyleMusic: A cross-modal generative framework for stylized AI music composition. Journal of Neural Audio Processing, 17 (2), 110– 125.
Ignatova, D. (2025). Application of wellness program for the prevention of aggressive behavior. Scientific Journal "Kindergarten School", 6 (1), 60–116.
Jain, M., & Aggarwal, R. (2024). Understanding and detecting deepfakes across visual and audio modalities. International Journal of Artificial Intelligence Research, (1), 78–99.
Mlynar, V., Polák, M., & Kruliš, M. (2023). DeepTextMark: A deep learning-driven text watermarking approach. Applied Sciences, 13 (17), 9875. https://www.mdpi.com/2076-3417/13/17/9875
Nguyen, T. T., Nguyen, Q. V. H., Nguyen, D. T., Nguyen, D. T., Huynh-The, T., Nahavandi, S., Nguyen, T. T., Pham, Q.-V., & Nguyen C. M. (2019) ‘ Deep Learning for Deepfakes Creation and Detection: A Survey ’. Computer Vision and Image Understanding, 223, Article 103525. https://doi.org/10.1016/j.cviu.2022.103525.
Ranjan, S., Li, H., & Das, T. (2023). DetectGPT and the future of synthetic text forensics. In Proceedings of the Computational Linguistics and AI Ethics Symposium (pp. 87–101).
Sabatelli, M., Sutherland, M., Cakmakci, O., & Reinecke, K. (2023). Organic or diffused: Can we distinguish human art from AI-generated images? In Proceedings of the 28th International Conference on Intelligent User Interfaces (IUI '23). https://dl.acm.org/doi/abs/10.1145/3576915.3623189
Solaiman, I., Clark, J., & Zhang, J. (2024). Human vs. machine: A comparative study on the detection of AI-generated content. arXiv Preprint, arXiv:2402.03214. https://arxiv.org/abs/2402.03214
Vasudeva, N., Kapoor, A., & Iyer, S. (2024). DeepTextMark: Watermarking-based authorship verification for AI-generated text. In Advances in AI for Digital Humanities (pp. 54–69). Springer.
Wolter, M., Blanke, F., Heese, R. & Garcke, J. (2022) Wavelet-packets for deepfake image analysis and detection. Machine Learning, 111, 4295–4327. https://doi.org/10.1007/s10994-022-06225-5
Zhou, L., He, K., & Yuan, M. (2025). UC-VITS: Cross-lingual voice cloning and emotional speech synthesis via variational inference. Transactions on Speech Technologies, 29 (1), 15–37.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2025 Digital Presentation and Preservation of Cultural and Scientific Heritage
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
