Last modified 5 August 2026

Artificial Intelligence in South Korea

High-risk AI in South Korea

Change countries

Change country

Law / proposed law in South Korea

On 26 December 2024, the ‘Framework Act on the Development of Artificial Intelligence and the Establishment of Foundation for Reliability’ (AI Act) passed the plenary session of the National Assembly. The AI Act was promulgated on 21 January 2025 (Law No. 20676) and came into effect on 22 January 2026. On 31 December 2025, the National Assembly passed an amendment (Law No. 21311, promulgated on 20 January 2026) that, among other matters, renamed the ‘National AI Committee’ (NAIC) to the ‘National AI Strategy Committee’ (NAISC), expanded its membership and mandate, introduced a definition of ‘training data’, established a legal basis for AI research institutes, and strengthened protections for AI-vulnerable groups in impact assessments. The Enforcement Decree (No. 36053) was promulgated on 21 January 2026 and came into effect together with the AI Act on 22 January 2026. On 20 July 2026, the Enforcement Decree (No. 36506) was amended to stipulate the details of the amendments to the AI Act.

The AI Act is intended to advance AI development and promote self-regulation by establishing a framework of the following initiatives: (i) formulating a master plan for AI by the Minister of the Ministry of Science and ICT (MSIT), creating the National AI Strategy Committee (NAISC, previously the ‘National AI Committee’) under the President’s office, establishing the AI Policy Center, and establishing the legal foundation for the AI Safety Institute’s operations; (ii) supporting industries related to the development and promotion of AI technology, including establishing standards for AI technology, defining ‘training data’ used in AI development and utilisation, and enabling universities and enterprises to establish AI research institutes with MSIT approval; and (iii) enacting and announcing the ‘AI Ethics Principles’ to support self-verification and certification by AI-related organisations, thereby ensuring the safety and reliability of AI and establishing the legal basis for autonomous ethics committees in the private sector.

Furthermore, the AI Act stipulates various obligations for AI business operators, such as operators involved with high-impact AI, businesses offering generative AI products or services, and operators whose AI systems exceed designated training compute thresholds, utilise cutting-edge technology and present a risk level capable of significantly impacting safety and fundamental rights of users (i.e., high-performance AI). It also requires that operators lacking a domicile or business location within Korea must appoint a domestic agent to comply with the regulatory framework and empowers the Minister of MSIT with the authority to conduct fact-finding inspections and to issue suspension or corrective orders where necessary.

The AI Act is the first statute to govern legal requirements specific to AI technologies and products in Korea.

Please note that AI in South Korea will still be regulated by existing rules governing personal information, copyright, and telecommunications. Therefore, the existing obligations and requirements under these laws and regulations will continue to apply to AI-related business and services.

Related resources

Regulatory guidance / voluntary codes

Regulatory guidance / voluntary codes in South Korea

Various governmental authorities such as MSIT, the Personal Information Protection Commission (PIPC) and the Korea Media Communications Commission (KMCC) (formerly Korea Communications Commission) are creating regulatory guidance. Following the AI Act’s commencement on 22 January 2026, the National Information Society Agency (NIA) published a suite of enforcement guidance documents on 22 January 2026, including practical guidelines for AI business operators on high-impact AI, generative AI, and domestic agent requirements. For general guidance on AI, the following can be considered:

‘National Guidelines for AI Ethics’ were prepared by MSIT in 2020, to provide comprehensive standards that should be followed by all members of society to implement ‘human-centred AI’. The National Guidelines for AI Ethics highlight three basic principles that should be considered during the development and utilisation of AI to achieve ‘AI for humanity’: (1) respect for human dignity; (2) the common good of society; and (3) proper use of technology. They also list ten key requirements that should be met throughout the AI system lifecycle to abide by the three basic principles, including safeguarding human rights, protection of privacy, prevention of harm, transparency, respect for diversity, and accountability, among others.

‘AI Ethics Self-Checklists’ were also prepared by MSIT and the Korea Information Society Development Institute (KISDI) in 2023 to help AI actors examine their adherence to the National Guidelines for AI Ethics in practice. They cover philosophical and social disclosures, including ethical considerations concerning the development and utilisation of AI, as well as social norms and values to be pursued. The AI Ethics Self-Checklists provide both a general-purpose checklist and a field-specific checklist that can be used by different AI actors, with the latter covering fields of AI chatbot, AI for writing, and AI image recognition systems.

Eight ‘Guidebooks for Development of Trustworthy AI’ were prepared by MSIT and the Telecommunications Technology Association (TTA) in 2023 and 2024, providing development requirements and verification items to be used as reference materials for ensuring trustworthiness in the process of developing AI products and services. The eight sector-specific versions of the Guidebooks for Development of Trustworthy AI provide sector-specific specialised use cases based on requirements and assessment questions of the general version of the same to enhance practical use. The sector-specific versions recommend selecting appropriate sector-specific requirements and assessment questions considering the characteristics of AI services during AI trustworthiness assurance activities, covering the medical, autonomous driving, public and social, general AI, smart security, and hiring sectors.

A ‘Strategy to Realize Artificial Intelligence Trustworthy for Everyone’ was also announced by the MSIT in 2021. It seeks to realise trustworthy AI for everyone applying three pillars (of technology, system and ethics) across ten action plans.

Related resources

Appointed supervisory authority

Appointed supervisory authority in South Korea

There is currently no single primary supervisory authority under the AI Act in Korea. The Minister of MSIT has overall responsibility for AI policy coordination, enforcement of the AI Act, and operation of the AI Safety Institute. However, it is expected that regulatory agencies will continue to handle specific issues according to their respective domains. For example, PIPC will handle personal information-related issues, KMCC communication regulation issues, and the Korea Fair Trade Commission (KFTC) fair trade issues. The National AI Strategy Committee (NAISC), chaired by the President and comprising ministers and private-sector experts, provides high-level policy deliberation and coordination.

Related resources

Definitions

Definitions in South Korea

Under the AI Act, the term ‘artificial intelligence’ or ‘AI’ is defined as the electronic manifestation of human intellectual capabilities such as learning, inference, perception, judgement, and language comprehension.

’AI system’, the principal subject of regulation, is defined as an AI-based system with varying degrees of autonomy and adaptability, capable of influencing physical or virtual environments through its predictions, recommendations, and decisions.

Furthermore, the AI Act applies to AI business operators, defined as individuals or entities engaged in AI-related activities. These operators are categorised into two groups (Article 2, Item 7): (i) ‘AI Developers’ (corporations, organisations, individuals, and national institutions involved in the development and provision of AI); and (ii) ‘AI Use Businesses’ (corporations, organisations, individuals, and national institutions that offer AI products or services using AI developed by others).

Prohibited activities

Prohibited activities in South Korea

The AI Act does not specifically enumerate or stipulate any specific prohibited actions (in contrast to the treatment of prohibited AI practices under the EU AI Act). However, actions that are already prohibited under existing laws and regulations, such as infringement of copyright or privacy, and distribution and publication of illegal information and contents, may still be problematic in relation to AI-related services.

High-risk AI

High-risk AI in South Korea

The AI Act outlines several key obligations for AI business operators who aim to provide high-impact AI systems or products or services utilising such technology.

High-Impact AI Definition: ‘High-Impact AI’ systems are those that significantly influence or pose risks to the safety and fundamental rights of individuals. These are typically employed in critical decision-making or assessments with substantial impact on someone’s rights and responsibilities. The AI Act (Article 2, Item 4) lists specific domains including: energy supply, drinking water production, healthcare systems, medical devices and digital medical products, nuclear safety, use of biometric information for investigation or arrests, recruitment and loan assessments that materially affect individuals’ rights and obligations, and key operation of transportation means, facilities, and systems. Additional domains may be designated by Enforcement
Preliminary Review Obligation: AI business operators must assess whether their AI technology qualifies as high-impact before deployment. They may seek confirmation from the Minister of MSIT if there is uncertainty regarding the classification of their AI system (Article 33).
Advance Notification Obligation: AI business operators intending to deploy products or services using high-impact AI are obligated to inform users in advance (Article 31, Paragraph (1)). Non-compliance may result in an administrative fine of up to KRW 30 million (Article 43, Paragraph (1), Item 1).
Safety and Reliability Measures: A comprehensive framework of safety and reliability measures must be implemented by operators offering high-impact AI systems to ensure these systems operate as intended without undue risk (Article 34).
Impact Assessment Obligation: AI business operators must use their efforts to assess the potential impact of their high-impact AI on individuals’ fundamental rights. Public institutions, including national and local government entities, must prioritise AI solutions that have undergone such assessments (Article 35).
Right to Explanation: Individuals affected by AI systems including high-impact AI have the right to request clear explanations of the logic and principles behind AI-generated outcomes, to the extent that this is technically and reasonably feasible (Article 3, Paragraph (2)).

Related resources

Controls on generative AI

Controls on generative AI in South Korea

The AI Act mandates several obligations on AI business operators that intend to offer products or services utilising generative AI.

Definition of Generative AI: This term refers to AI systems that produce content such as text, audio, images, and other outputs by mimicking the structure of input data (Article 2, Item 5).
Advance Notification Obligation: AI business operators must notify users in advance that their products or services are powered by generative AI (Article 31, Paragraph (1)). Non-compliance may result in an administrative fine of up to KRW 30 million (Article 43, Paragraph (1), Item 1).
Labelling Obligation: Products or services must be clearly labelled as being created by generative AI (Article 31, Paragraph (2)).
Deepfake Content: AI business operators providing virtual outputs that may be mistaken for real (often referred to as ‘deepfakes’), must ensure these are clearly labelled. If labelled content qualifies as artistic or creative expression, the manner of labelling should not hinder its appreciation (Article 31, Paragraph (3)).
Compliance Guidance: The specifics of notification and labelling, including potential exceptions, are detailed in the Enforcement Decree (No. 36053), which came into effect on 22 January 2026. The NIA also published practical guidance on generative AI labelling requirements in January 2026.

Enforcement / fines

Enforcement / fines in South Korea

The AI Act provides authority to the Minister of MSIT to initiate investigations in cases where (i) MSIT learns of any actual or potential violation of the following obligations under the AI Act, or (ii) MSIT receives a report or civil complaint of such a violation: (i) obligation to label content created using generative AIs (Article 31, Paragraph (2)); (ii) obligation to provide notice to viewers or to label ‘deepfakes’ (Article 31, Paragraph (3)); (iii) obligation to secure safety of high-performance AIs and/or duty to report on measures taken by the AI business operators to secure such safety of high-performance AIs (Articles 32, Paragraphs (1) and (2)); and (iv) obligation to secure safety and reliability for high-impact AIs (Article 34, Paragraph (1)). Upon finding of any violation listed above, the Minister of MSIT may issue an order to suspend or correct the action in violation against the violator (Article 40, Paragraph (3)).

Furthermore, administrative fines may be imposed for the following: (i) failure to appoint a domestic agent may result in an administrative fine of up to KRW 30 million (Article 43, Paragraph (1), Item 2); (ii) failure to comply with the advance notification obligation for the high-impact AI or generative AI may result in an administrative fine of up to KRW 30 million (Article 43, Paragraph (1), Item 1); and (iii) failure to comply with the corrective orders may result in administrative fines of up to KRW 30 million (Article 43, Paragraph (1), Item 3).

Related resources

User transparency

User transparency in South Korea

Certain notification, labelling and/or explanation obligations are required for high-impact AI and generative AI, as discussed in the High-Risk Uses and Controls on Generative AI sections.

Fairness / unlawful bias

Fairness / unlawful bias in South Korea

Currently, the AI Act does not clearly stipulate this, but it is recommended in the National Guidelines for AI Ethics and other similar documents referred to in the Regulatory Guidance / Voluntary Codes section.

Human oversight

Human oversight in South Korea

The AI Act does not specifically mandate human oversight as a general obligation. However, it does require certain safety and reliability measures in relation to high-impact AI. For high-performance AI systems, the AI Act requires AI business operators to identify, assess, and mitigate risks throughout the AI life cycle, and establish a risk management system (Article 32). For high-impact AIs, human oversight and supervision is required as one of the Safety and Reliability Measures (Article 34, Paragraph (1), Item 4). The Enforcement Decree (No. 36053) specifies certain procedures for high-impact AI confirmation requests and safety/reliability disclosures. Additionally, the AI Safety Institute (established under Article 12 of the AI Act) is tasked with research into AI safety and may contribute to the development of specific human oversight standards in due course.

Related resources

Key contacts

Key contacts in South Korea

Taeyoung Roh

Attorney

Kim & Chang

Seoul

Email Full bio

Michael Kim

Senior Foreign Attorney

Kim & Chang

Seoul

Email Full bio

The AI Act outlines several key obligations for AI business operators who aim to provide high-impact AI systems or products or services utilising such technology.

High-Impact AI Definition: ‘High-Impact AI’ systems are those that significantly influence or pose risks to the safety and fundamental rights of individuals. These are typically employed in critical decision-making or assessments with substantial impact on someone’s rights and responsibilities. The AI Act (Article 2, Item 4) lists specific domains including: energy supply, drinking water production, healthcare systems, medical devices and digital medical products, nuclear safety, use of biometric information for investigation or arrests, recruitment and loan assessments that materially affect individuals’ rights and obligations, and key operation of transportation means, facilities, and systems. Additional domains may be designated by Enforcement
Preliminary Review Obligation: AI business operators must assess whether their AI technology qualifies as high-impact before deployment. They may seek confirmation from the Minister of MSIT if there is uncertainty regarding the classification of their AI system (Article 33).
Advance Notification Obligation: AI business operators intending to deploy products or services using high-impact AI are obligated to inform users in advance (Article 31, Paragraph (1)). Non-compliance may result in an administrative fine of up to KRW 30 million (Article 43, Paragraph (1), Item 1).
Safety and Reliability Measures: A comprehensive framework of safety and reliability measures must be implemented by operators offering high-impact AI systems to ensure these systems operate as intended without undue risk (Article 34).
Impact Assessment Obligation: AI business operators must use their efforts to assess the potential impact of their high-impact AI on individuals’ fundamental rights. Public institutions, including national and local government entities, must prioritise AI solutions that have undergone such assessments (Article 35).
Right to Explanation: Individuals affected by AI systems including high-impact AI have the right to request clear explanations of the logic and principles behind AI-generated outcomes, to the extent that this is technically and reasonably feasible (Article 3, Paragraph (2)).

Quick links

Artificial Intelligence in South Korea

High-risk AI in South Korea

Continue reading