Seoul, Jan 22 (NationPress) On Thursday, South Korea officially implemented a comprehensive law for the safe utilization of artificial intelligence (AI) models, becoming the first country in the world to do so. This pioneering legislation establishes a regulatory framework aimed at combating misinformation and mitigating other risks associated with this rapidly evolving field. The Basic Act on the Development of Artificial Intelligence and the Establishment of a Foundation for Trustworthiness, commonly referred to as the AI Basic Act, became effective on Thursday, as confirmed by the science ministry, according to reports from Yonhap news agency.

This significant move marks the initial governmental adoption of extensive guidelines regarding AI usage worldwide.

The law emphasizes the need for companies and AI developers to assume greater accountability in addressing deepfake content and misinformation generated by AI systems, empowering the government to impose penalties or initiate investigations for any violations.

Specifically, the act introduces the designation of “high-risk AI,” which pertains to AI models that can profoundly impact users' daily lives or safety, including applications in the hiring process, loan applications, and medical consultations.

Entities utilizing such high-risk AI models are mandated to inform users that their services are AI-driven and are responsible for ensuring user safety. Furthermore, any content produced by these AI models must carry watermarks that indicate its AI-generated origin.

“Implementing watermarks on AI-generated content serves as a fundamental measure to prevent the misuse of AI technology, such as deepfake scenarios,” stated a ministry official.

International companies providing AI services in South Korea that meet certain criteria—such as an annual global revenue of 1 trillion won (approximately $681 million), domestic sales of 10 billion won or more, or a user base of at least 1 million daily users in the country—are required to appoint a local representative.

Currently, OpenAI and Google meet these criteria.

Any violations of the act may incur fines up to 30 million won, with the government planning to implement a one-year adjustment period before penalties are enforced to assist the private sector in adapting to the new regulations.

In addition, the act features provisions for the government to foster the AI industry, necessitating the science minister to present a policy blueprint every three years.