🧔♂️ A friendly human may check it before it goes live. More news here
S Korea to require labels on AI-generated ads from Jan 2026
South Korea will require clear labels on AI-generated online content starting January 2026, aiming to reduce fake and misleading ads.
The new rules come as AI-generated ads featuring fake experts and celebrity deepfakes have risen sharply, especially in the food and pharmaceutical sectors.
The Ministry of Food and Drug Safety reported nearly 97,000 AI-generated ads for food and drug products in 2024, up from about 59,000 in 2021.
Under the policy, anyone producing, editing, or uploading AI-generated photos or videos must disclose their origin, and deleting or altering labels is banned.
Online platforms must ensure content creators follow the labeling rules, and provide standardized methods and notifications.
Ads using AI-generated doctors or experts to promote products will be considered deceptive if not clearly marked as virtual humans.
Regulators will also introduce stricter penalties, including punitive damages up to five times the actual harm, and plan to revise related laws in 2026.
🔗 Source: The Korea Times
🧠 Food for thought
Implications, context, and why it matters.
Korea’s AI labeling mandate adds granular obligations to global transparency rules
- EU AI Act plus China’s 2023 measures require AI content transparency. Korea’s January 2026 rules also assign labels to producers, editors, and uploaders across the full content lifecycle [News Article]1.
- Korea Communications Commission (KCC) guidance in 2025 recommends visible watermarks, digital markers such as AI logos. The AI Framework Act, a national law setting broad AI rules, requires labels for synthetic content “indistinguishable from reality”. The new policy bans label removal, which creates enforceable standards 1[News Article].
- The policy answers a real misuse problem. AI-made ads with fake experts jumped 64% from 2021 to 2024, hitting nearly 97,000 food-drug spots in 2024, and platform rules failed to curb it [News Article].
- Oversight spans product advertising run by the Ministry of Food and Drug Safety (MFDS) plus user-protection guidance for platforms from the KCC 1[News Article]. The U.S. rolled back prior federal AI-labeling guidance in early 2025, so the national approach remains mixed.
Compliance software providers can target Korea’s major ad platforms ahead of 2026 enforcement
- Naver and Kakao lead in local services. YouTube reaches about 43.4 million users, or 91%, in 2025 within a multibillion-dollar ad market 23.
- Vendors can ship provenance labels that use Coalition for Content Provenance and Authenticity (C2PA) metadata with watermarking for these platforms [News Article]2. Penalties can reach five times actual damages under enforcement [News Article]2.
- The USD 6.5 billion online ad market needs automated compliance checks and label-verification APIs, given high AI ad volumes in 2024 [News Article]3.
- Startups that detect deepfakes, by spotting AI-altered faces or voices, can offer real-time blocks before posts go live for roughly 7.18 million TikTok users and 23.6 million Instagram users 2.
Stay updated on the go with our mobile app.
Get latest insights with smoother, more personalized experience through TIA mobile app.