Naver has officially begun a closed beta test for its next-generation search service, the 'AI Tab', signaling a major strategic shift towards licensed data.
This move is driven by two powerful trends sweeping the tech industry. First is the intense global competition in AI search, where companies like Google and Perplexity are racing to innovate. Second is a critical shift away from indiscriminately scraping web data for AI training, a practice now fraught with legal challenges. Instead, the new standard is to secure high-quality, rights-cleared data through licensing agreements, which is precisely what Naver is doing with partners like EBS (Korea Educational Broadcasting System) and Doosan Encyclopedia.
Several factors have led Naver to this critical juncture. Firstly, the company faces mounting legal pressure. An ongoing lawsuit from major broadcasters over unauthorized use of news content for AI training has made it clear that relying on unlicensed data is a significant business risk. Securing clean data from EBS and Doosan directly mitigates this threat. Secondly, Naver has the necessary infrastructure. The recent completion of a powerful 4,000-GPU Nvidia B200 cluster provides the computational power needed to train its AI, HyperCLOVA X, on these vast new datasets. Thirdly, competitive threats, such as Google Maps potentially gaining fuller functionality in Korea, are pushing Naver to strengthen its unique, localized advantages.
By integrating verified content, Naver aims to significantly improve the reliability of its AI-generated answers and reduce instances of 'hallucination'—when an AI confidently presents false information. The data from EBS offers high-quality educational material, while Doosan provides a deep well of curated general knowledge. This not only enhances the user experience but also reinforces Naver's moat as the most trusted search portal in Korea.
This strategic pivot could also unlock shareholder value. Currently, Naver's stock trades at a price-to-earnings (P/E) ratio well below its historical average. If the AI Tab successfully increases user engagement and monetization while lowering legal risks, it could justify a re-rating of the stock, presenting a potential upside for investors. Ultimately, Naver's success in the new AI era will likely depend less on the raw power of its algorithms and more on the quality and integrity of its data.
- Hallucination: An event where an AI model generates incorrect or nonsensical information but presents it as factual.
- P/E (Price-to-Earnings) Ratio: A valuation metric that compares a company's current share price to its per-share earnings. A lower P/E can sometimes suggest a stock is undervalued.
- Moat: A term used in business to describe a company's sustainable competitive advantage that protects it from competitors, much like a moat protects a castle.