Can ChatGPT Detect DeepFakes? A Study of Using Multimodal Large Language Models for Media Forensics

Shan Jia, Reilin Lyu, Kangran Zhao, Yize Chen, Zhiyuan Yan, Yan Ju, Chuanbo Hu, Xin Li, Baoyuan Wu, Siwei Lyu; Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2024, pp. 4324-4333

Abstract

DeepFakes which refer to AI-generated media content have become an increasing concern due to their use as a means for disinformation. Detecting DeepFakes is currently solved with programmed machine learning algorithms. In this work we investigate the capabilities of multimodal large language models (LLMs) in DeepFake detection. We conducted qualitative and quantitative experiments to demonstrate multimodal LLMs and show that they can expose AI-generated images through careful experimental design and prompt engineering. This is interesting considering that LLMs are not inherently tailored for media forensic tasks and the process does not require programming. We discuss the limitations of multimodal LLMs for these tasks and suggest possible improvements.

Related Material

[pdf] [arXiv]
[bibtex]
@InProceedings{Jia_2024_CVPR, author = {Jia, Shan and Lyu, Reilin and Zhao, Kangran and Chen, Yize and Yan, Zhiyuan and Ju, Yan and Hu, Chuanbo and Li, Xin and Wu, Baoyuan and Lyu, Siwei}, title = {Can ChatGPT Detect DeepFakes? A Study of Using Multimodal Large Language Models for Media Forensics}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2024}, pages = {4324-4333} }