Public Notary System in Korea: Document Authentication Guide (EN/中文/日本語/Español)

🌐 Language / 語言 / 言語 / Idioma: English 中文 日本語 Español 📋 Korean Law Insight | Korean Legal System By Attorney NOH UJIN | HANBYOL LAW LLC, Seoul Public Notary (공증) System in Korea: A Complete Legal Guide to Document Authentication for Foreign Individuals and Businesses Korea's notarization system — operating under the Notary Public Act (공증인법) and the Civil Procedure Code's provisions on notarial deeds — differs significantly from common-law notary systems familiar to many foreign businesspeople and individuals. Understanding how Korean public notaries (공증인) function is essential for anyone executing powers of attorney, loan agreements, corporate resolutions, or other documents intended for use in Korea or abroad. 1. What Is a Korean Public Notary, and How Does It Differ from the Western Notary? In common-law jurisdictions (US, UK, etc.), a "notary public" is typically a low-level official who verifies identity and witnesses signatures. I...

South Korea's New AI Framework Act: A Compliance Guide for Foreign Tech Companies

# South Korea's New AI Framework Act: A Compliance Guide for Foreign Tech Companies The global race for artificial intelligence (AI) leadership is accelerating, and with it, the critical need for robust regulatory frameworks. South Korea, a global innovation powerhouse, has taken a significant step forward with the enactment of its **Act on the Development of Artificial Intelligence and Establishment of Trust (인공지능 발전과 신뢰 기반 조성 등에 관한 기본법)**. Set to take effect in January 2026, this landmark legislation aims to foster a dynamic AI ecosystem while ensuring ethical development, user protection, and societal trustworthiness. For foreign tech companies operating or planning to enter the vibrant Korean AI market, understanding and navigating this new legal landscape is not just a recommendation but a fundamental prerequisite for success. ## Contextualizing Korea's AI Regulatory Ambition The `AI 기본법` (AI Framework Act) represents Korea's proactive approach to balancing the immense potential of AI with its inherent risks. Its core purpose, as stated in **Article 1 (Purpose of Enactment)**, is to "protect the Rights and Interests of the People and the dignity of the People, to improve the quality of life of the people, and to strengthen national competitiveness by prescribing the basic matters necessary for the sound development of AI and the establishment of a foundation for trustworthiness." This dual objective underscores a commitment to both innovation-driven growth and responsible AI governance. Foreign tech companies must recognize that their operations in Korea will be evaluated not only on technological merit but also on their adherence to these foundational principles. ## Decoding Key Provisions for Compliance The `AI 기본법` introduces several critical concepts and mandates that foreign tech companies must meticulously incorporate into their compliance strategies. ### Core Principles and Ethical Foundations The Act emphasizes several guiding principles that permeate its provisions. While specific articles elaborate on these, the overall spirit aligns with human-centered AI, transparency, accountability, safety, reliability, accessibility, and contribution to human well-being. **Article 27 (AI Ethical Principles)**, for instance, mandates the government to establish and publish AI Ethical Principles to ensure AI development and utilization does not cause harm to human life, physical well-being, or mental health, promotes accessibility for all, and contributes to human well-being and prosperity. Companies are expected to align their AI development and deployment with these national ethical standards. ### Defining the Scope: Key Terminology **Article 2 (Definitions)** is crucial for understanding the Act's applicability. It defines key terms, including: * **"Artificial intelligence"**: The electronic implementation of human intellectual abilities such as learning, reasoning, perception, judgment, and language comprehension. * **"Artificial intelligence system"**: An AI-based system inferring results that affect real and virtual environments with autonomy. * **"High-Impact AI"**: AI systems that may significantly affect or pose risks to life, physical safety, or basic rights in designated domains (e.g., energy supply, healthcare, criminal investigation with biometric info, employment/loan decisions, public service decision-making). This classification triggers more stringent compliance obligations. * **"Generative AI"**: AI systems that generate text, sound, images, video, or other content by mimicking input data structures. * **"AI Business Operator"**: Entities engaged in the AI industry, including AI Developers and AI-Using Business Operators. This broad definition ensures comprehensive coverage. * **"Affected Person"**: An individual whose life, physical safety, or basic rights are significantly affected by AI Products or AI Services. ### Enhanced Transparency for High-Impact and Generative AI One of the most direct compliance requirements falls under **Article 31 (Duty to Ensure AI Transparency)**. AI business operators providing products or services utilizing High-Impact AI or Generative AI are mandated to: * **Notify users in advance** that such products or services are AI-operated (**Article 31, Paragraph 1**). * **Clearly indicate to users** that outputs are generated by Generative AI (**Article 31, Paragraph 2**). * For AI systems generating virtual audio, images, or video indistinguishable from real content, **notify or indicate to users** in a clearly recognizable manner that the content is AI-generated (**Article 31, Paragraph 3**). This ensures users are aware when interacting with AI or consuming AI-generated content, fostering trust and preventing deception. ### Robust Safety Measures **Article 32 (Duty to Ensure AI Safety)** introduces obligations for AI business operators whose AI systems meet or exceed a certain cumulative computing capacity (to be prescribed by Presidential Decree). These operators must: * **Identify, assess, and mitigate risks** throughout the entire lifecycle of the AI system (**Article 32, Paragraph 1, Subparagraph 1**). * **Establish and maintain a risk management framework** capable of monitoring and responding to AI safety incidents (**Article 32, Paragraph 1, Subparagraph 2**). * **Submit the results of these safety measures** to the Minister of Science and ICT (**Article 32, Paragraph 2**). Furthermore, for **High-Impact AI**, **Article 34 (Measures to Ensure Safety and Trustworthiness of High-Impact AI)** requires additional measures, including: * Establishing and operating a **risk management plan** (**Article 34, Paragraph 1, Subparagraph 1**). * Formulating and implementing measures to provide **explanations regarding final AI outputs, principal criteria used, and an overview of learning data** (**Article 34, Paragraph 1, Subparagraph 2**). This is a critical explainability requirement. * Establishing and operating **user protection measures** (**Article 34, Paragraph 1, Subparagraph 3**). * Ensuring **human oversight** of High-Impact AI operations (**Article 34, Paragraph 1, Subparagraph 4**). * **Preparing and retaining documentation** demonstrating compliance (**Article 34, Paragraph 1, Subparagraph 5**). ### Data Handling and the Personal Information Protection Act While the `AI 기본법` sets out the framework for AI, the bedrock of data handling in South Korea remains the **Personal Information Protection Act (개인정보보호법)**. Any AI system processing personal information must fully comply with PIPA, which mandates strict requirements for: * **Consent**: Obtaining explicit consent for the collection, use, and provision of personal information (e.g., **Article 17 (Provision of Personal Information to Third Parties)**, **Article 18 (Matters Requiring Consent of Information Subject)**). * **Anonymization/Pseudonymization**: Utilizing anonymized or pseudonymized data wherever possible, with specific guidelines for processing such data (**Article 58-2 (Special Cases concerning Processing of Pseudonymized Information)**, **Article 58-3 (Special Cases concerning Processing of Anonymized Information)**). * **Security**: Implementing technical, administrative, and physical measures to secure personal information (**Article 29 (Duty to Take Measures for Securing Personal Information)**). * **Cross-border Data Transfers**: Adhering to strict conditions for transferring personal information outside of Korea (**Article 39-8 (Transfer of Personal Information Abroad)**). Foreign tech companies developing or deploying AI in Korea must integrate PIPA compliance into their data governance strategies for AI, as violations can lead to severe penalties. ### Impact Assessments for Fundamental Human Rights **Article 35 (Impact Assessment for Fundamental Human Rights)** encourages AI business operators providing High-Impact AI to "endeavor, in advance, to conduct an assessment of the potential impact on fundamental human rights." While framed as an endeavor, national institutions are mandated to prioritize High-Impact AI products or services that have undergone such an assessment. This signals a strong regulatory expectation for proactive ethical consideration and could become a de facto requirement for competitive market entry. ### The Domestic Representative Requirement for Foreign Entities A particularly salient provision for foreign tech companies is **Article 36 (Designation of Domestic Representative by Overseas AI Business Operator)**. Foreign AI business operators without a domestic address or place of business in South Korea, whose user numbers or revenue meet criteria prescribed by Presidential Decree, *must designate a domestic representative*. This representative is responsible for: * Submitting results of safety measure implementation (**Article 32, Paragraph 2**). * Requesting confirmation on High-Impact AI qualification (**Article 33, Paragraph 1**). * Supporting the implementation of safety and reliability measures (**Article 34, Paragraph 1**). This mirrors similar requirements in data protection laws globally and ensures a clear point of contact and accountability within Korea for compliance with the AI Framework Act. ## Takeaway: Proactive Compliance is Key South Korea's `AI 기본법` signifies a mature and comprehensive approach to AI regulation. For foreign tech companies, success in this advanced market hinges on a deep understanding of these laws and a proactive approach to compliance. Simply translating existing global compliance frameworks will not suffice; a localized strategy that respects Korea's unique legal and ethical nuances is essential. Investing in legal counsel with expertise in Korean AI and data protection law is not merely an expense but a strategic investment to ensure smooth operations, foster trust, and avoid significant legal and reputational risks in one of the world's leading AI economies. By Attorney NOH UJIN | HANBYOL LAW LLC, Seoul Email: ujin95@hanbl.co.kr

🌐 | | |

Comments

Popular posts from this blog

Taxi, Car-Sharing and Platform Transportation Regulation in Korea: A Legal Guide for Foreign Mobility Companies (EN/中文/日本語/Español)

How Foreign Creditors Can Recover Debts in South Korea: A Step-by-Step Legal Guide

Commercial Arbitration in South Korea: What Foreign Businesses Must Know