Photo courtesy of Yonhap News

 

[Alpha Biz= Paul Lee] SEOUL, April 9, 2026 — Naver asserted in court that not all news content qualifies as fully protected copyrighted material, as it faces a lawsuit from major South Korean broadcasters over the alleged unauthorized use of news data in training its generative AI model, HyperCLOVA X.

The fourth hearing in the copyright infringement injunction case was held on April 9 at the Seoul Central District Court, where terrestrial broadcasters KBS, MBC, and SBS argue that Naver used their news content without permission during the development of its AI system.

During the hearing, Naver cited Article 7 of South Korea’s Copyright Act, which excludes certain categories — including laws, official documents, and factual news reporting intended solely to convey information — from copyright protection. The company maintained that its argument was limited to the scope of permissible use for AI training and did not imply that all news content lacks protection.

A Naver representative stated that “current affairs reporting may fall under the category of factual information not subject to copyright,” adding that the company’s use of data is permitted under existing news partnership agreements.

The broadcasters strongly rejected this claim, emphasizing that news content is a refined intellectual product created through extensive reporting, editing, and editorial expertise accumulated over decades. They argued that such content carries significant economic value and credibility and should be fully protected.

They also pointed out that Choi Soo-yeon previously described news content as “high-quality data essential for AI training” during the launch of HyperCLOVA X, criticizing Naver’s current stance as undermining the value of journalism. The broadcasters further disputed Naver’s claim that partnership agreements grant blanket rights to use news articles for AI training without compensation.

The Korea Broadcasters Association warned that the case could set a precedent for how AI companies in South Korea approach copyright, expressing concern that Naver’s position may discourage licensing negotiations and content purchases across the industry.

The lawsuit is widely seen as a key test case in defining the boundaries of copyright and data usage for AI training between digital platforms and media companies.

Separately, Naver announced that it will discontinue its Korean-language chatbot service Clova X — built on HyperCLOVA X — effective April 9, 2026. The company said it plans to continue expanding the application of HyperCLOVA X across broader industrial domains.

 

 

[ⓒ AlphaBIZ. 무단전재-재배포 금지]