en.Wedoany.com Reported - South Korea's Personal Information Protection Commission (PIPC) has released the "Roadmap for Technology Development and Standardization of Full Lifecycle Protection and Utilization of Personal Information (2026~2030)", aimed at addressing the risks of personal information leakage and misuse in the artificial intelligence environment and establishing a trust-based data utilization system.
This roadmap integrates and connects the previously separate technology development and standardization roadmaps, making revisions in advance to enhance continuity between technology development and standardization, while proactively responding to technological changes in the AI field. With the proliferation of new technologies such as agent AI and physical AI, the personal information processing environment is undergoing rapid changes, raising concerns about information leakage and re-identification of AI training data. The roadmap reorganizes the protection and utilization technologies and standardization targets that need to be studied across the entire lifecycle of personal information, from generation and collection to destruction.
The roadmap comprehensively considers domestic and international technology and standardization trends, defines a classification system for full lifecycle protection and utilization technologies of personal information, and selects 11 core technologies across four major areas. The four major areas include safeguarding personal information sovereignty, mitigating leakage/exposure risks, trust-based secure utilization, and AI response technology development. Among the 11 core technologies, new technical elements capable of addressing the AI environment have been incorporated, such as security assessments for information leakage risks that may occur during AI model training and inference, and technologies to prevent misuse and abuse of personal information in agent AI and physical AI environments.
Specific core technologies include: policy compliance verification result inquiry in safeguarding personal information sovereignty; deepfake/synthesis verification and labeling, edge device personal information protection, and dark web/surface web leakage detection in mitigating leakage/exposure risks; re-identification risk assessment and verification, privacy-enhancing technologies (PET) based on synthetic data, de-identification (single/hybrid), and MyData consent and delegation integrated automation platform in trust-based secure utilization; AI model security assessment, agent, tool, and robot execution security, physical AI real-time privacy control, and AI-based unstructured data de-identification (images, text, voice, etc.) in AI response technology development.
Through research on AI-PET technologies integrating personal information protection enhancement with AI, and unstructured data de-identification technologies, the plan aims to promote practical and efficient technology development and standardization that balances personal information protection and secure utilization in an environment where various new technologies emerge. Additionally, the roadmap newly establishes a talent cultivation direction for the next decade to secure professionals with capabilities in personal information protection and utilization, as well as leakage accident prevention and response.
This article is compiled by Wedoany. All AI citations must indicate the source as "Wedoany". If there is any infringement or other issues, please notify us promptly, and we will modify or delete it accordingly. Email: news@wedoany.com
