South Korea passes world’s first comprehensive AI law, reshaping global regulation

South Korea has enacted the world’s first comprehensive set of laws to regulate artificial intelligence and strengthen the transparency, compliance and safety in the sector.

Dubbed as AI Basic Act, the law passed on Thursday will position the East Asian country as the leader in global governance of artificial intelligence.

The law has taken effect sooner than expected contrary to EU AI Act, which is being applied in phases through 2027.

“The law will provide a critical institutional foundation" for South Korea's ambition to become a top-three global AI powerhouse, Science minister Bae Kyung-hoon, a former head of AI research at electronics giant LG, told a press conference.

However, the enactment of regulatory laws has sparked the concerns among the tech firms and startups, arguing that compliance will curtail their potential in the tech-driven world.

Key postulates of AI Basic Act

The law focuses on two main things: safety and honesty.

The companies must be obliged to ensure human supervision in “high-impact” AI fields, including nuclear safety, healthcare, transport, and finances.

Moreover, the law also requires companies to give their users advance and clear notice about products and services that use generative AI, along with transparent labeling.

According to the Ministry of Science and ICT, the regulatory framework is designed to promote AI adoption while building a foundation of trust and safety.

To help businesses and startups adjust, companies must be given at least one year to follow rules before the government starts imposing hefty fines.

In the case of non-compliance and violation, the companies could be fined up to 30 million won ($20,400).

Evoking resentment among tech firms

According to Lim Jung-wook, co-head of South Korea's Startup Alliance, the key details have caused resentment among the businesses.

“There's a bit of resentment — why do we have to be the first to do this?” he said.

However, the ministry has affirmed to support the companies in navigating the regulatory landscape.

"Additionally, we will continue to review measures to minimise the burden on industry," a spokesperson said.

AI global governance

When it comes to AI global governance, the world remains divided on how to regulate artificial intelligence.

The US, being a key player, favours a more light-touch, sectoral, and decentralized governance approach to avoid stifling innovation. In 2025, America’s AI Action Plan was introduced based on “light-touch” reality.

On the contrary, China promotes a robust and state-centric governance framework as some experts fear that it will hinder innovation and AI advancements. China has passed several vertical laws, including

Algorithm Recommendation Provisions (2021), Generative AI Measures (2023), and AI Security Governance Framework 2.0 (2025)

Last year in November, China proposed a global body to govern artificial intelligence and enact uniform rules, making AI a "public good for the international community".

If the dilemma regarding the absence of uniform AI governance persists, the artificial intelligence which is meant to be for greater good would become the victim of geopolitics and zero-sum mentality.