The Personal Information Protection Commission allocated 1.296 billion won in the first supplementary budget this year to strengthen the detection system for synthetic and edited materials such as deepfakes.
According to the commission on the 4th, this budget will be used to enhance the detection system for personal information leaks and exposures, reflecting the changing nature of illegal distribution forms from text-based to unstructured data. Unstructured data refers to information such as voice, video, and images that do not have a defined structure.
In the case of deepfakes, which emerged as a social issue last year, the illegal synthesis of individuals' faces and personal information into images or videos showed limitations in filtering them out with existing detection systems.
In response, the commission decided to expand the detection targets from text to focus on images and videos, and to incorporate artificial intelligence (AI) methods, including machine learning technology, into the detection system that was previously operated based on keywords.
A commission official noted, “We will establish a system that can identify personal information included in images or videos through the approved supplementary budget, minimizing personal information leaks and illegal distribution and preventing secondary damage.”