S. Korea launches world’s first AI law: Bold move or premature gamble?

Описание: 세계 첫 AI 기본법 시행… 선도 전략인가, 시기상조인가

Welcome to Within The Frame, where we bring the most pressing issues across the globe into focus. I'm Kim Mok-yeon.
Starting today, Korea becomes the first country in the world to enforce a national AI law.
The AI Basic Act introduces legal standards for transparency, safety, and accountability, while aiming to promote growth in the AI industry.
Unlike the EU, which delayed its AI regulation to 2027, Korea moves ahead, with rules for high-impact AI, mandatory labeling of AI-generated content, and a one-year grace period for companies to adjust.
But the law is already facing concerns over vague definitions, possible overreach, and its impact on startups and content creators.

So what does this law actually change—starting now?

For answers to this, we connect to Song Se-ryun, Professor of Law at Kyung Hee University. Welcome.

Also joining us is Park Juyong, a Professor at the Graduate School of Culture Technology at KAIST. Good to see you.


(PARK) 1. So let's begin with Prof.Park, today, Korea became the first country in the world to fully enforce a national AI law.
What is the significance of being first, and what kind of global responsibility or expectation comes with that?

(SONG) 2. Now that the law took effect, people are wondering what they’ll actually notice.
Prof.Song, what changes immediately under this law, and what should the public understand as the most visible difference starting today?

(PARK) 3. One of the most noticeable changes is the new rule on transparency—mandatory labeling for AI-generated content, especially deepfakes.
From a technology and content perspective Prof.Park, how significant is this labeling system, and how realistically can services comply across different platforms?

(SONG) 4.That brings us to one of the law’s most debated concepts, the category of “high-impact AI.”
Legally speaking Prof.Song, what qualifies as high-impact AI under this framework?

(PARK) 5. To break that down further, the law considers factors like autonomy, risk level, and domain of use when classifying high-impact AI.
Prof.Park, do you think these criteria are practically applicable to real-world AI systems?

(SONG) 6. And naturally, this classification raises concerns for many in the industry—especially smaller players.
Prof.Song, What are the most pressing legal or structural concerns you see with how the high-impact AI category is being rolled out?

(PARK) 7. Let’s stay with the practical implications—especially for content creators.
The law allows lighter disclosure rules for internal platform use like chatbots, but tightens them for anything shared externally.
Prof.Park, Does this two-tiered approach strike the right balance between protecting users and encouraging creative freedom?

(SONG) 8. That also raises the issue of legal responsibility.
The law puts the burden on AI service providers—not users—when it comes to safety and transparency.
Prof.Song, Is that legally appropriate, especially in cases involving misuse or harmful content like misinformation or deepfakes?

(SONG) 9. This law also applies to foreign AI firms serving Korean users, like Google or OpenAI.
Can Korea realistically enforce these provisions against global Big Tech Prof.Song, and how does this affect domestic firms competing under the same law?

(PARK) 10. Finally, stepping back, This law is being positioned as the foundation for Korea’s global AI leadership.
In the long run, how might this law shape Korea’s AI ecosystem, its public trust in tech, and its place in the global AI race Prof.Park?


We'll have to leave our discussion right here, many thanks to Prof.Song and Park. we appreciate it.

#SouthKorea #AI #Law #AI_law #Bold #Gamble #AI_Guardrails #SongSeryun #ParkJuyong #대한민국 #인공지능 #법 #AI_기본법 #시행 #Arirang_News #아리랑뉴스

📣 Facebook :   / arirangtvnews  
📣 Twitter :   / arirangtvnews  
📣 Homepage : https://v2.arirang.com/

2026-01-22, 20:30 (KST)