Metacloud is an AI security corporations that centers its technology on "AI content trust (Trust Technology)." Chief Executive Kim Hyeong-jin founded it in Feb. 2022. Kim has served as a professor in the Future Convergence Management Department at Myongji University, researching deepfake detection and verification of AI-generated content.

Kim Hyeong-jin, CEO of Metacloud

Kim said, "I decided to start a company because I believed research results should not remain only in papers but be implemented as technologies that work in the field."

Kim clearly distinguished between the existing security market and today's changes. "If security in the past was about protecting systems, now the key is judging whether the content we see and hear is real or fake," Kim said, explaining that "deepfakes and AI-generated content are spreading as primary tools for financial fraud, identity theft, and fake news."

Market growth is also steep. The global deepfake detection market is projected to grow by about 40% annually through 2034. The AI content verification market is also expected to expand to $7.4 billion (about 10.88 trillion won) by 2031. "Threats are increasing rapidly, but corporations' responses are severely lacking," Kim said, calling it "a textbook demand-gap market."

The regulatory environment also supports market expansion. "The European Union's AI Act has mandated labeling for AI-generated content starting in 2025," Kim said. "With similar regulations being introduced in Korea, the United States, and China, content verification is becoming essential infrastructure." Kim added, "In the end, this market is a competition over 'who can most accurately distinguish the real from the fake.'"

Metacloud has built a technology portfolio that detects Generative AI content across voice, text, images, and video. In particular, it has raised trustworthiness by combining watermarking with detection technology. Its flagship service, "That Guy's Voice," launched in Nov. 2025. "That Guy's Voice analyzes real-time call content on smartphones with on-device AI, regardless of carrier or device model," Kim said, describing it as "a dual detection solution that performs both deep voice detection and voice phishing intent analysis."

Metacloud's AI voice phishing detection service Geunom Voice./Courtesy of Metacloud

Kim cited "speed and architecture" as the technological edge. "Real-time detection is possible in 5–10 second intervals, achieving speeds two to three times faster than before," Kim said. "This enables preemptive response before damage occurs." Kim added, "Because it is 100% On-device AI, there are no concerns about personal data leaks, and it works even in areas with poor network conditions."

Technologies in images and video are also nearing commercialization. "We have completed development of a video call module that can detect advanced deepfakes such as face swaps," Kim said, adding, "The voice model will be expanded into AI Contact Center (AICC) applications to improve both consultation efficiency and expense structure."

Global expansion is also in full swing. Metacloud supports Arabic based on a model trained in Korean and English simultaneously. "We launched and are operating That Guy's Voice in Singapore in Feb. 2026," Kim said. "Based on this, we plan to expand rapidly into North America and Southeast Asia."

A text-based detection solution, "Geuljaguk (working title)," is also in the works. "We extended the technologies validated in voice to the text domain," Kim said. "Simultaneous support for Korean and English will be a global competitive edge."

"Our goal is to evolve into an 'AI content trust platform' that spans voice, text, and video," Kim said, emphasizing, "We will establish ourselves not as a single-solution corporations but as a technology-based corporations for the Generative AI era."

※ This article has been translated by AI. Share your feedback here.