The Personal Information Protection Commission (hereinafter referred to as the Personal Information Commission) announced on the 19th that it has released a "model for AI privacy risk management for safe artificial intelligence (AI) and data utilization" (hereinafter referred to as the risk management model) to provide systematic guidelines for corporations to autonomously manage privacy risks.
The risk management model systematically organizes the direction and principles of privacy risk management throughout the AI lifecycle, the types of risks, and mitigation measures, reflecting the need to respond to the advancements in AI technology and the increase in personal data breaches and new risks.
In particular, the Personal Information Commission developed the risk management model through discussions of the Risk Assessment Task Force of the public-private policy council since December of last year. This model comprehensively incorporates characteristics of AI data processing, types of privacy risks, and mitigation measures discussed by domestic and international academia, government, and research institutions.
The risk management model recommends periodic management starting from the planning and development stages of AI based on procedures such as ▲ identifying AI types and use cases ▲ risk identification and measurement ▲ establishing safety measures. Additionally, it presents specific measures to mitigate risks through administrative and technical safety measures.
The Personal Information Commission plans to continuously update the risk management model considering advancements in AI technology and global trends, and will also prepare additional materials specialized for small corporations and specific areas.
Professor Park Sang-cheol of Seoul National University, who led this discussion, noted, "This is a meaningful model aimed at systematically addressing the gap between rapidly advancing AI technology and personal information protection regulations." Background Hoon Bae, head of LG Research Institute, said, "We have laid the foundation for our country to play a leading role in international discussions on AI privacy."
Chairperson Koh Hak-soo of the Personal Information Commission said, "Given the uncertainties in the AI domain, reasonable and proportional management is necessary," and added, "I hope this risk management model will provide substantial assistance in managing privacy risks for corporations in the AI sector."