What You See Is Not What You Know: Deepfake Image Manipulation

What You See Is Not What You Know: Deepfake Image Manipulation
What You See Is Not What You Know: Deepfake Image Manipulation
File Size:
737.34 kB
Author:
Cathryn Allen, Bryson R. Payne, Tamirat T. Abegaz, Chuck Robertson
Date:
31 December 2022
Downloads:
55 x
Research indicates that deceitful videos tend to spread rapidly online and influence people's opinions and ideas. Because of this, video misinformation via deepfake video manipulation poses a significant online threat. This study aims to discover what factors can influence viewers' capability of distinguishing deepfake videos from genuine video footage. This work focuses on exploring deepfake videos' potential use for deception and misinformation by exploring people's ability to determine whether videos are deep fakes in a survey consisting of deepfake videos and original unedited videos. The participants viewed a set of four videos and were asked to judge whether the videos shown were deepfakes or originals. The survey varied the familiarity that the viewers had with the subjects of the videos. Also, the number of videos shown at one time was manipulated. This survey showed that familiarity with the subject(s) depicted in a deepfake video has a statistically significant impact on how well people can determine it is a deepfake. Notably, however, almost two-thirds of study participants (102 out of 154, or 66.23%) were unable to correctly identify a sequence of just four videos as either genuine or deepfake. The potential for deepfakes to confuse or misinform a majority of the public via social media should not be underestimated. This study provides insights into possible methods for countering disinformation and deception resulting from the misuse of deepfakes. Familiarity with the target individual depicted in a deepfake video contributed to viewers' accuracy in distinguishing a deepfake better than showing unaltered authentic source videos side-by-side with the deepfakes. Organizations, governments, and individuals seeking to contain or counter deepfake deception will need to consider two main factors in their operational planning: 1) a swift, near-real-time response to deepfake disinformation videos, and 2) creating more familiarity through additional, preferably live video footage of the target of the deepfake responding to and refuting the disinformation personally.
 
 
Powered by Phoca Download