Recent Regulatory Trends in AI Industry

In November 2025, the Korean government announced the “AI Regulatory Rationalization Roadmap” (the “AI Roadmap”) to lay out the foundation for enhancing national competitiveness in the AI industry. In January 2026, the Framework Act on the Development of Artificial Intelligence and Establishment of Trust (the “AI Basic Act”) and its Enforcement Decree officially took effect, and key guidelines were finalized and published. These developments are driving rapid changes in Korea’s AI regulatory environment.

AI Roadmap

The AI Roadmap announced by the Korean government on November 27, 2025 is the government’s first roadmap for rationalizing regulations in emerging industries. Corporations are encouraged to carefully review each initiative of the AI Roadmap, as doing so will help business operators gauge the direction and focus of the Korean government’s core AI policies.

The AI Roadmap identifies four major segments of the AI industry, which are (i) technology development, (ii) service utilization, (iii) infrastructure, and (iv) reliability and security. It also specifies key issues for each segment and provides regulatory rationalization measures for each issue as follows:

Segment	Key Regulatory Issues	Regulatory Measures and Improvement Plans
Technology Development	Securing and utilizing AI training data	Establish the “Guidelines on Fair Use for AI Training,” specifying criteria and providing examples for determining fair use under the Copyright Act and consider amendments to relevant statutes after soliciting feedback from stakeholders Support transactions regarding AI training data by providing a negotiation framework, etc.
	Protecting and utilizing AI-generated outputs	Develop examination criteria and application guidelines for industrial property rights (including patent and design rights) relating to AI-generated outputs, and review amendments to the relevant regulations
	Infrastructure for various data utilization	Revise various laws/regulations and publish guidelines to promote the utilization of industrial and manufacturing data, synthetic data and pseudonymized data
	Expansion of the use of public data	Expand the scope of disclosure of public data by designating and promoting high-value public datasets based on business demand
Service Utilization	Mobility and intelligent robot development	Expand the scope of demonstrations of autonomous driving, permit the use of original video data in autonomous-driving development, and ease regulations for the use of outdoor mobile robots
Service Utilization	Improvement of administrative efficiency	Improve administrative efficiency by deploying AI in public services
Infrastructure	Data centers	Improve uniform regulatory requirements, such as mandatory installation of artwork and elevators in data centers
Reliability and Security	Revision of the AI Basic Act	Establish subordinate statutes, notifications and guidelines related to the criteria for identifying high-impact AI under the AI Basic Act and detailed implementation measures

As the regulatory issues and initiatives outlined above are likely to take shape in the future through legal and regulatory reforms and the establishment of guidelines, AI-related companies and investors should carefully analyze how these changes will potentially affect their businesses and services and prepare accordingly.

Enforcement of AI Basic Act and Publication of Subordinate Guidelines

The AI Basic Act took effect on January 22, 2026, and its Enforcement Decree—together with five guidelines (available in Korean, Link) outlining the key obligations of AI deployers under the AI Basic Act—was also finalized and published.

Enforcement Decree of AI Basic Act

The revised Enforcement Decree of the AI Basic Act, which took effect concurrently with the AI Basic Act, provides more detailed guidance compared to the initial draft bill for the Enforcement Decree that was pre-announced in November 2025. Notably, the revised Enforcement Decree clarifies the category of AI business operators required to designate a domestic agent as “AI deployers who have been subject to administrative fines for failing to comply with suspension and/or corrective orders issued by the Ministry of Science and ICT (“MSIT”).”

Subordinate Guidelines of AI Basic Act

Meanwhile, the AI Basic Act guidelines, initially released in September 2025, have been finalized after collecting feedback from stakeholders for approximately four months. Among them, the AI Transparency Guidelines relating to Article 31 of the AI Basic Act underwent the most substantial revisions, as follows:

Scope of the obligation to ensure transparency: Clearly provide that the obligation to ensure transparency applies strictly to “AI deployers” and not “users.”

Differentiation of labeling obligations by service type: Distinguish between (i) AI-generated outputs provided only within a service environment, and (ii) AI-generated outputs exported outside a service environment through downloading or sharing, while allowing for more flexible labeling in the former case.

In addition, the MSIT announced its plan to launch a system improvement task force beginning in February 2026 to continuously review potential improvements and revisions to the AI Basic Act and the applicable guidelines. Therefore, it will be necessary to take into account future amendments and changes to the guidelines.

Grace Period for Enforcement under AI Basic Act

The MSIT has announced its plans to implement a grace period of at least one year under which enforcement of the AI Basic Act would be suspended, to allow companies to adapt and minimize disruption. During this grace period, fact-finding investigations and the imposition of administrative fines are expected to occur only in exceptional circumstances, such as where serious social issues arise (e.g., fatal accidents or human rights violations).

Further, as the MSIT has indicated its plan to collect feedback from stakeholders during the grace period to ensure the reasonable operation of the regulatory system, AI business operators should closely follow related developments.

Various legal frameworks in the AI sector will continue to be revised in line with the publicly announced AI Roadmap, and the guidelines and enforcement direction of the AI Basic Act are also expected to change following broad consultation with the public during the grace period. To maintain flexibility, business operators in the relevant sectors should assess the potential risks of their services based on the currently available legal frameworks and keep track of regulatory developments and updates to the AI Basic Act, which will be specified according to the AI Roadmap.

[Korean Version]

Share

Professionals

Professionals