South Korea is set to enforce a comprehensive AI regulatory framework on January 22, 2026, becoming the world's first country to do so. The law aims to promote AI development through a national committee and three-year plans while imposing safety and transparency requirements, but startups express concerns over insufficient preparation time and burdens on smaller firms.
South Korea's National Assembly passed the Framework Act on the Development of Artificial Intelligence during a plenary session on November 27. The law is scheduled to take effect on January 22, 2026, mandating the establishment of a national AI committee, the creation of a basic three-year AI plan, and safety and transparency requirements, including disclosure obligations for certain AI systems.
If implemented as planned, South Korea would be the first country worldwide to enforce a comprehensive AI regulatory framework. The European Union passed AI-related legislation first but plans to apply most rules starting in August, with some provisions delayed until 2027 due to business pressures and intensifying global competition.
However, startups and businesses in South Korea have raised concerns about the timeline. An official from the Korea Internet Corporations Association stated, "Companies may not have sufficient time to prepare for the new rules, as the enforcement decree is expected to be finalized only shortly before the law takes effect due to procedural requirements." He added, "This will be particularly overwhelming for startups."
A recent survey by Startup Alliance revealed that 98 percent of 101 local AI startups have not established a compliance system for the new law. Among respondents, 48.5 percent were unfamiliar with the law and unprepared, while another 48.5 percent were aware but ill-equipped.
Another industry official warned, "If the current implementation timeline is maintained, some companies may be forced to abruptly change or suspend services after Jan. 22." He continued, "If regulations are too strict, companies will have stronger incentives to launch services overseas rather than at home."
Such regulatory pressures are seen as a factor driving more South Korean AI startups to consider Japan, which adopts a softer, voluntary governance approach. Concerns are particularly high over mandatory watermarking rules for AI-generated content, aimed at curbing deepfakes and misuse.
An official at an AI content company noted, "Even AI-generated contents often involve hundreds of people working to improve quality, but consumers may turn away once they are labeled as 'AI-generated.'" He added, "There are also ambiguities in the labeling requirements, and I don't think the views of those who fully understand the content creation industry and other relevant experts have been fully reflected."
