Deputy Prime Minister Bae Kyunghoon Visits AI Safety Research Institute: "We Will Create an AI Environment the Public Can Trust"

Deepfake and AI Agent Era: Industry-Academia-Research Roundtable
Comprehensive National AI Safety Ecosystem Plan to Be Established Within the Year

On October 23, Deputy Prime Minister and Minister of Science and ICT Bae Kyunghoon visited the Artificial Intelligence Safety Research Institute to inspect the site of artificial intelligence (AI) safety technology and held a roundtable with experts from industry, academia, and research institutes. With a recent surge in crimes using deepfakes and growing concerns that unpredictable damages may occur in the era of Artificial General Intelligence (AGI) and AI agents, the meeting was held to discuss the current status of AI safety technologies and policy directions.

The AI Safety Research Institute was established in 2024 and is responsible for research on AI safety technologies, model evaluation, and global cooperation to address national-level AI risks. This aligns with the global trend of major countries such as the United Kingdom (November 2023), the United States (February 2024), Japan (February 2024), and Singapore (May 2024) establishing similar institutions.

On this day, Deputy Prime Minister Bae visited the institute's AI research evaluation site and encouraged the research team working on key projects, including ▲deepfake detection, ▲future AI risk research, ▲AI safety evaluation, and ▲AI safety policy and international cooperation. He also observed demonstrations of deepfake detection technology and AI agent safety evaluation, and discussed ways to ensure AI safety with 15 experts from industry, academia, and research institutes.

The subsequent roundtable featured two sessions under the theme "Current Status of Technology Development for AI Risk Response and Strategies for Building an AI Safety Ecosystem." The first session focused on the development direction of technologies to prevent deepfakes and ensure the reliability of AI agents, while the second session concentrated on the implementation of a (tentative) "National AI Safety Ecosystem Comprehensive Plan" in consideration of global AI risk response trends. Participants agreed on the need to build an ecosystem that enables the development and use of safe and trustworthy AI.

Deputy Prime Minister Bae stated, "AI that the public can use with confidence is the first step toward a society where AI and humans coexist," and added, "We will work with relevant ministries and agencies to establish a 'National AI Safety Ecosystem Comprehensive Plan' within this year so that AI safety can be integrated into all industries and daily life."