Naver Is Paying Creators to Feed Its AI. What Should You Do?

Key Points

• Naver is committing 1 trillion won (~$670 million) over five years to pay content creators whose work fuels its AI search summaries.
• User-generated content accounts for about 70% of sources cited in Naver's AI Briefing feature, making creator content the actual product powering AI search.
• This is the same structural bet Google made with its $60 million Reddit deal: AI companies need quality training data more than they need bigger models.
• Publishers outside Korea should read this as a signal, not a curiosity. The question of who pays for content, and how, is moving fast.
• Your traffic may already be feeding AI products. Whether you're getting compensated for it depends entirely on what decisions you make now.

South Korea's dominant search engine just made the most honest statement in AI right now. According to Tech Times, Naver announced a 1 trillion won (~$670 million) investment fund to pay creators whose content trains its AI Briefing product. The company's Chief Data and Content Officer said the quiet part loud: as AI model performance levels off, the value of high-quality data outweighs sheer model scale.

That's not a niche Korean market story. That's the entire AI content debate compressed into a single corporate strategy.

What Naver Is Doing

Naver's plan is more structured than a typical creator fund. The program, called Naver Mate, selects 3,000 creators monthly from its blog, community, Q&A, and premium content services. Selection criteria are based on how often a creator's work gets cited in AI Briefing across 10 content categories.

The pay structure breaks down like this:

The fund distributes 20 billion won annually across those 3,000 monthly slots. The selection signal matters: citation frequency in AI summaries determines who gets paid. Naver is, in effect, telling creators exactly what makes their content valuable to an AI system.

User-generated content already accounts for about 70% of sources cited in AI Briefing. Naver isn't building this infrastructure because it's generous. It's building it because it needs that content to keep improving its product, and it can't rely on passive scraping forever.

See It In Action:

• Entertainment Content Website Case Study: How an entertainment publisher maximized ad revenue and navigated a shifting traffic environment with Playwire's full-stack platform.
• Lifestyle, Health & Travel Ad Revenue Resource Center: Publisher-specific strategies for maintaining revenue performance as organic and AI-driven traffic patterns evolve.

Why This Matters to Publishers Outside Korea

The Naver story is the Reddit deal at scale. Google pays Reddit roughly $60 million a year for access to its content for AI training. Naver is building an internal version of that arrangement with its own creator base. The structural logic is identical in both cases: AI products need high-quality, human-generated content, and the people who produce it want compensation.

For publishers outside Naver's ecosystem, two things are worth understanding.

First, your content is almost certainly being used to train or feed AI products right now. Whether you're being paid for it is a separate question, and for most publishers, the answer is no. Naver is making explicit what the rest of the industry is doing implicitly.

Second, the citation-based selection model Naver uses reveals where AI search value actually accumulates. It's not going to the publisher who produces the most content. It's going to the publisher whose content gets cited most often in AI-generated summaries driving publisher traffic. Volume doesn't win this race. Authority does.

Essential Background Reading:

• AI Crawler Resource Center for Publishers: The full reference hub for understanding how AI crawlers interact with your site and what decisions you need to make.
• AI and Publishers Resource Center: Playwire's complete collection of publisher guidance on navigating the AI search transition.
• Generative AI and Publishers: An overview of how generative AI is reshaping the publishing landscape and what it means for ad revenue.
• AI Overview for Publishers: Core concepts and context for publishers getting up to speed on AI's impact across ad tech and content strategy.

What Publishers Should Consider Now

The Naver model is a preview, not a blueprint you can copy directly. It does clarify some decisions worth making, though.

Here are the practical questions this story should prompt:

• Your robots.txt configuration: Are you actively deciding which AI crawlers can access your content, or are you operating on defaults you set three years ago? This is a business decision, not a technical formality.
• Your content positioning: AI Briefing cites sources based on authority and relevance. The publishers who get cited are the ones who will eventually get visibility. If your content isn't structured to be cited, you're producing training data, not traffic.
• Your compensation strategy: Naver's model pays creators explicitly. Outside that ecosystem, compensation comes indirectly through traffic, brand authority, or licensing. If traffic from AI referrals is dropping, the other two options need to replace it.
• Your revenue concentration: If display advertising is your primary monetization layer, AI-driven traffic shifts hit you harder than publishers with diversified revenue. This is the time to stress-test your monetization stack.

The AI search transition isn't arriving. It's happening. Naver's 30 million monthly AI Briefing users and the 3 million who picked up AI Tab in its first month show you how fast adoption moves once a product reaches scale.

Related Content:

• Should You Block AI Crawlers? A decision framework for publishers weighing the tradeoffs of blocking AI bots versus allowing access.
• AI Crawler Protection Grader: Audit your current crawler configuration and see how well your site is protected against unauthorized AI scraping.
• AI Content and Publisher Rights: What publishers need to understand about how their content is being used by AI systems and what protections exist.
• The Quality vs. Quantity Revolution: Why ad load and traffic stability matter more than raw volume in a shifting search environment.

The Data vs. Model Argument

Naver's Chief Data Officer made a specific claim worth sitting with: model performance is leveling off, and data quality is now the differentiator. A Seoul National University researcher, quoted in the source reporting, framed Naver's real asset as two decades of localized data. The open question, per that researcher, is whether Naver's AI can become "good enough, fast enough" to convert that data advantage into actual product wins against OpenAI and Google.

That framing applies to publishers, too. Your archive, your audience data, your first-party signals: these are the assets that AI companies are competing for through licensing deals. Whether you treat them as leverage or give them away passively is a strategic choice with long-term revenue consequences.

Next Steps:

• Future-Proofing Your Content Strategy: A practical guide to deciding whether blocking AI crawlers fits your long-term content and revenue strategy.
• AI Traffic Is the New SEO: How publishers can position their content to capture referral traffic from AI-generated search summaries.
• Reader Revenue Manager: The Complete Publisher's Guide: How identity, engagement data, and first-party signals translate into stronger ad revenue regardless of traffic source.
• Complete Guide to Ad Monetization for Publishers: Diversify your monetization stack so revenue doesn't depend entirely on traffic volume from any single source.

The Revenue Picture for Publishers

Naver's model points to one path: a platform pays you directly when your content serves its AI product. Outside walled gardens like Naver, that arrangement doesn't exist for most publishers yet. What does exist is the ad revenue sitting on every page your audience actually visits.

If AI search is reducing your referral traffic, the sessions you retain become more valuable, not less. Maximizing revenue per session on existing traffic isn't a consolation strategy. It's the right operational response to a traffic environment that's getting harder to predict.

We work with publishers across gaming, entertainment, education, and news who are navigating exactly this. Tighter traffic means tighter tolerance for underperforming ad setups. Our RAMP platform is built to squeeze the most out of every session, with yield ops expertise, full auction transparency, and a demand stack that's actually competitive. If your current setup isn't performing against that standard, we should talk.