State of AI Search Russia 2026: The Complete AI Visibility Report
GEO Scout flagship report: 10 AI providers, 716 brands, 5 niches of the Russian market. Full analysis of AI search in Russia for 2026 — content intent, cited sources, entry paths, diversity, and niche benchmarks.
In 2025, Athena HQ published "State of AI Search 2026" — the first large-scale AI search report based on 8 million responses from the Western market. It changed the global conversation around GEO. But the Russian market was nearly absent from it.
Russia has its own AI ecosystem. YandexGPT and Alice AI reach 88 million users who do not use ChatGPT as their primary AI tool. Google AI Mode exited beta specifically in early 2026. DeepSeek saw explosive growth among Russian audiences following its February model update. Copilot is embedded in Microsoft 365 and has occupied the corporate AI search niche.
At GEO Scout we monitor 10 AI providers daily and have accumulated data that exists nowhere else. This report is our "State of AI Search Russia 2026": a complete picture of brand AI visibility in the Russian market — from content intent to entry paths and diversity of sources.
Executive Summary: 10 Key Findings
-
AI traffic has reached 15% and keeps growing — according to GEO Scout client data, the share of AI-source referrals to commercial websites reached 15% in Q1 2026. Forecast for end of 2026: 20%+.
-
Google AI Mode became the #1 provider for Russian brands — Timeweb received an average position of 1.43, Skillbox hit 100% Mention Rate. Years of SEO investment are now converting into AI visibility through the Google index.
-
YandexGPT + Alice AI — a unique local duet — Yandex's two platforms together reach 88 million monthly active users. No Western provider has comparable coverage of the Russian-speaking audience.
-
Gemini knows the Russian market better than ChatGPT — in EdTech, e-commerce and hosting, Gemini delivers 100% Mention Rate for top brands. ChatGPT, by contrast, performs well only in e-commerce and largely ignores Russian EdTech.
-
Aviasales is the absolute recommendation leader — 36.6% combined Recommendation Rate, 76.7% in Gemini, 70% in Grok. No other Russian brand receives so many direct AI recommendations.
-
Community signals are underutilized — Habr, vc.ru, Pikabu, DTF, and Yandex.Zen are actively cited by AI providers, yet brands do not work with these platforms systematically. This is the biggest unclosed gap.
-
Content Intent has 9 categories, not 4 — AI search processes intent types fundamentally differently from SEO. Informational, comparative, transactional and other types produce different visibility patterns across different providers.
-
Entry Paths: /home beats /blog in tourism and e-commerce — in tourism and commerce queries, AI providers more often link to the home page and product pages than to blog content. Half of brands do not know this.
-
Diversity: Perplexity is twice as diverse as Copilot — 14-17 domains versus 8-10, meaning different entry strategies for each provider.
-
DeepSeek — high domain citation despite average reach — in EdTech, DeepSeek places a link to netology.ru in every second response (55.2% domain citation). This is direct AI traffic without advertising spend.
Methodology: How We Measure
All data in this report was collected by the GEO Scout platform during Q1 2026 (January — March).
| Parameter | Value |
|---|---|
| Period | January — March 2026 |
| AI providers | 10: ChatGPT, Claude, DeepSeek, Gemini, Google AI Mode, Google AI Overview, Grok, Perplexity, YandexGPT, Alice AI |
| Niches | 5: E-commerce, FinTech, EdTech, Hosting, Tourism |
| Brands in database | 716+ |
| Responses analyzed | 12,000+ over the quarter |
| Monitoring frequency | Daily for each provider |
Key metrics in this report:
- Mention Rate — percentage of AI responses where the brand is mentioned at least once
- Share of Voice (SoV) — brand's share of all mentions within its niche (sum = 100%)
- Avg Position — average position in a list when AI names multiple brands (1.0 = always first)
- Recommendation Rate — percentage of responses with a direct recommendation ("I recommend", "best option")
- Domain Citation Rate — how often AI places a link to the brand's domain
- Positive Rate — share of positive mentions out of all mentions
Section 1: AI Search Landscape — 10 Providers in the Russian Market
The Russian AI search market differs fundamentally from the Western one: two fully local providers (YandexGPT and Alice AI) exist here with a combined audience of 88 million users, and Google AI Mode only exited Russian-language beta at the start of 2026.
Provider Map: Knowledge of the Russian Market
| Provider | Russian Market Knowledge | Notes |
|---|---|---|
| YandexGPT | Excellent | Local provider, Yandex ecosystem |
| Alice AI | Excellent | Second Yandex local provider, 88M MAU |
| Gemini | High | 100% Mention Rate for top brands in 4 niches |
| Google AI Mode | High | Exited RU beta Q1 2026, best positional metrics |
| Claude | Moderate | "All or nothing": either 100% or 0% |
| Perplexity | Moderate | Consistent coverage, always shows sources |
| Grok | Moderate | Record recommendations, real-time index |
| DeepSeek | Moderate | High domain citation, variable reach |
| ChatGPT | Weak for RU | Strong only in e-commerce and international brands |
| Google AI Overview | Moderate | Search snippets, focus on informational queries |
Key Events in Q1 2026
Google AI Mode exited Russian-language beta in January 2026 — and immediately showed the best positional metrics of all providers. Timeweb received an average position of 1.43, Skillbox hit 100% Mention Rate. For all brands with strong SEO, Google AI Mode became an instant dividend: years of Google index optimization converted into AI visibility.
Grok integrated real-time search in Q1 and began recommending brands it had previously ignored. Aviasales received a 70% Recommendation Rate — a record for the tourism niche across all providers.
DeepSeek updated its model in February 2026 — positions of dozens of Russian brands shifted by 10-15%. This demonstrates the key risk of AI visibility: a provider can radically rewrite its "opinion" on brands after a single update.
YandexGPT + Alice AI continue to dominate the Russian-speaking audience. Alice AI now processes queries in Alice, Yandex Browser, and the smart device ecosystem — it is the AI search entry point for millions of users who have never opened ChatGPT.
Section 2: Content Intent in Russia — 9 Categories Instead of 4
Traditional SEO taxonomy divides queries into 4 types: informational, navigational, commercial, and transactional. AI search operates on a different logic.
According to Athena HQ for the Western market, informational content accounts for 27-49% depending on the vertical, comparative for 19-26%, and transactional for 11-17%. We observe a similar picture for the Russian market with important differences.
Content Intent Distribution for the Russian AI Market
| Intent type | Share of AI queries (RU) | Leading providers |
|---|---|---|
| Informational | ~35% | Google AI Overview, Perplexity |
| Comparative | ~24% | Google AI Mode, Gemini |
| Transactional | ~19% | YandexGPT, Alice AI |
| Navigational | ~9% | ChatGPT, Grok |
| Problem-solving | ~5% | Claude, Perplexity |
| Recommendational | ~4% | Grok, Google AI Mode |
| Reputational | ~2% | Gemini, Claude |
| Educational | ~1% | Claude, DeepSeek |
| Local/geo | ~1% | YandexGPT, Alice AI |
How Provider Type Affects Intent
Google AI Overview dominates informational queries — it appears above organic results for "what is...", "how does... work" questions. This means: in competitive niches, the informational AI Overview becomes a barrier between the user and organic results.
YandexGPT receives a high share of transactional queries — Yandex users ask "where to buy," "which bank is better," "how to order." This directly converts into commercial traffic for brands that appear there.
Grok with real-time search captures recommendational queries — "what should I choose right now," "current deals." This explains its record Recommendation Rates.
A full taxonomy of all nine Content Intent types with prompt examples and content strategies is detailed in the article «Content Intent Taxonomy for GEO».
Section 3: Top Cited Sources in Russia — Who Gets into AI Responses
One of Athena HQ's key findings: in the Western market, reddit.com dominates citations across all verticals (15-30%), with YouTube in a consistent second place (10-20%). In Russia the picture is fundamentally different.
Global Sources That Work in Russia
YouTube is the only global source that performs equally well in both hemispheres. According to our data, YouTube appears in 15-25% of AI responses that contain any links to external sources. This holds across all 10 providers — including YandexGPT and Alice AI.
Wikipedia is strong in informational queries, especially in Perplexity and Google AI Overview. Brand pages on Wikipedia regularly appear in AI responses to navigational and reputational queries.
Russian Sources in AI Citations
| Source | Content type | Niche strength |
|---|---|---|
| vc.ru | Expert articles, case studies | FinTech, E-commerce, SaaS |
| Habr | Technical reviews, tutorials | EdTech, hosting, technology |
| Yandex.Zen | User and author content | Broad reach, YandexGPT-affine |
| Pikabu | UGC, reviews, user experience | E-commerce, tourism, consumer |
| DTF | Games, media, technology | Young audience |
| RuTube | Video content | YouTube alternative for RU providers |
| Otzovik/IRecommend | User reviews | E-commerce, tourism |
What This Means for Content Strategy
If your brand wants to appear in AI responses, it is not enough to have a site blog. You need to be present where AI gets its information. In Q1 2026 we found: brands with an active presence on Habr and vc.ru receive 23-31% higher Domain Citation Rate in technical niches compared to brands that publish only on their own resources.
A full analysis of cited sources is in the article «Cited Sources in AI». The role of community signals is detailed in «Community Signals: Reddit, GitHub, Forums, and Habr».
Section 4: Diversity of Sources — How Many Domains Each AI Shows
Diversity is the number of unique domains that an AI provider mentions or cites in a single response. This metric directly affects how difficult it is to "enter" each provider.
Diversity Comparison by Provider
| Provider | Domains per response | Characteristic |
|---|---|---|
| Perplexity | 14-17 | Maximum diversification, always shows sources |
| Google AI Mode | 8-12 | High, relies on Google index |
| Claude | 1-5 | "All or nothing" — either knows you or doesn't |
| Gemini | 5-8 | Moderate, consistent coverage |
| ChatGPT | 3-6 | Moderate, rarely shows links |
| Grok | 4-8 | Growing with real-time integration |
| DeepSeek | 3-6 | Moderate but high domain citation |
| YandexGPT | 4-7 | Focus on Russian sources |
| Google AI Overview | 3-5 | Limited, focus on top-3 |
| Alice AI | 3-5 | Ecosystem-based, Yandex-affine |
A detailed diversity analysis by provider is in «How Many Domains Does Each AI Cite». Why Copilot behaves like an encyclopedia is covered in «Copilot — the Most Informational AI».
Section 5: Entry Paths — Blog vs Home vs Products in Russian Niches
Athena HQ identified an important pattern: /blog pages lead as the entry point in 6 of 8 Western verticals, but in retail and real estate the /home page dominates. We ran the same analysis for Russian niches.
Entry Paths by Russian Market Niche
| Niche | Primary Entry Path | Secondary Entry Path | Notes |
|---|---|---|---|
| EdTech | /blog, /courses | /home | Educational content dominates |
| E-commerce | /home, /catalog | /blog | Home page and catalog |
| FinTech | /home | /blog, /products | Home page as trust anchor |
| Tourism | /home, /search | /blog | Service pages beat content |
| Hosting | /blog | /pricing, /features | Technical content leads |
Why /home Matters More for Tourism and E-commerce
In tourism and e-commerce queries, AI providers understand that the user wants to do something right now — find a flight, compare prices, order a product. So AI more often links to the service home page or the search/catalog page rather than educational materials.
Example: "Find cheap flights" — AI links to aviasales.ru (home), not to an article "how to choose airline tickets." For Aviasales this means: the home page is the primary entry point for AI traffic.
For EdTech, the opposite holds. "How to choose an online Python course" — AI cites a review article in the blog, a course comparison, or a specific course landing page. /blog pages capture the majority of AI citations in the educational niche.
A full entry paths analysis with per-provider data is in «Entry Paths in AI Search: Blog vs Home vs Products».
Section 6: Benchmarks Across 5 Niches
Detailed per-niche data is in separate articles. Below is the consolidated comparison table with key metrics.
Consolidated Q1 2026 Benchmark
| Niche | Leader by Mention Rate | Position Record | Recommendation Record | Top Provider |
|---|---|---|---|---|
| E-commerce | Yandex.Market 97.05% | Wildberries 1.31 | Ozon — high | Gemini, Google AI Mode |
| FinTech | Alfa-Bank (SoV 12.53%) | T-Bank 1.89 | T-Bank (domain citation 26.7%) | Gemini, Google AI Mode |
| EdTech | Skillbox 100% (Claude) | Yandex Practicum 1.5 (Gemini) | — | Gemini, Google AI Mode |
| Tourism | Aviasales 97.05% | — | Aviasales 76.7% (Gemini) | Gemini, Grok |
| Hosting | Timeweb 100% (three providers) | Timeweb 1.43 (Google AI Mode) | Timeweb 53.3% (Grok) | Google AI Mode, Gemini |
E-commerce: The Unreachable Trio
The top-3 marketplaces receive 87-97% Mention Rate in most providers. This is practically the ceiling — AI always names them in any Russia-shopping query. The competition is for positions 4-5: Megamarket (46.7% in ChatGPT vs 100% in Gemini) and AliExpress (83.3% in ChatGPT) show significant variation between providers.
Full e-commerce brand ranking: «E-commerce Brand Rating in Neural Networks».
FinTech: Alfa-Bank Beats Sberbank in AI
One of the biggest surprises of Q1 2026 — Alfa-Bank with SoV 12.53% outranks Sberbank (10.81%) in the AI rating. AI forms responses based on content, not revenue or market share. Alfa-Bank is more actively represented in expert reviews, comparative articles, and fintech platforms — and AI sees that.
Full FinTech ranking: «FinTech Brand Rating in Neural Networks».
EdTech: Maximum Provider Divergence
EdTech is the niche with the highest spread between providers. SkillFactory: 6.7% in ChatGPT vs 72.4% in Claude. This means ChatGPT and Claude users see a fundamentally different online education market. For EdTech brands, the GEO strategy must be differentiated by provider.
Full EdTech ranking: «EdTech Brand Rating in Neural Networks».
Tourism: Aviasales Without Competition
Aviasales is the only Russian brand with a record combined Recommendation Rate (36.6%). In Gemini — 76.7%, in Grok — 70%, in Google AI Mode — 50%. This means: in 7 out of 10 Gemini responses to tourism queries, the AI directly recommends Aviasales. No competitor comes close.
Full tourism ranking: «Tourism Brand Rating in Neural Networks».
Hosting: Maximum Inter-Provider Spread
Hosting shows the widest per-provider spread for a single brand. Timeweb: 43.3% in ChatGPT vs 100% in Gemini, Google AI Mode, and Perplexity. AdminVPS: 0% in ChatGPT vs 62.9% in Claude. For hosting providers this means: GEO strategy must be targeted to specific providers, not to the market as a whole.
Full hosting ranking: «Hosting Brand Rating in Neural Networks». Benchmarks across all niches: «AI Visibility Benchmarks by Niche». Cross-niche findings: «5 Niches, 716 Brands: Key Conclusions».
Section 7: 2026 Forecast — What Will Change
1. AI Traffic Will Break the 20% Barrier
In Q1 2026, AI traffic at GEO Scout clients reached 15%. Our forecast: by the end of 2026 this figure will exceed 20% on commercial websites. Growth drivers: Google AI Mode scaling, AI integration in Yandex Browser, growing Perplexity audience, and Grok expansion with real-time search.
2. Google AI Mode Will Become a Mandatory Channel
After exiting beta, Google AI Mode already shows the best positional metrics of all providers. In 2026 it will become a mandatory channel for all brands with strong SEO. The "SEO and GEO in sync" principle will finally materialize in the data.
3. DeepSeek V4 Will Reshuffle the Rankings
The next DeepSeek update (forecast: Q3 2026) will again shift brand positions. Companies that monitor visibility daily will adapt within days. Those who check quarterly will lose months of visibility before the next review.
4. Grok with Real-Time Search Will Strengthen Recommendational Content
Grok already shows record Recommendation Rates (Aviasales 70%). As real-time search develops, it will increasingly recommend brands with current offers — discounts, promotions, fresh ratings. This is a new content category for GEO: "current offer right now."
5. YandexGPT 5 and Alice Integration
Yandex has announced the next generation of YandexGPT. Integration with Alice AI and smart devices means AI search will reach audiences that never thought about AI assistants. Voice queries through Alice are a new frontier for local brands — especially in FMCG and tourism.
6. Multimodal AI Search
Google AI Mode and ChatGPT already support images in queries. By end of 2026 we expect growth in multimodal AI queries: a user photographs a product and asks "where to buy cheaper" or "who makes this in Russia." For e-commerce this is a radically new type of traffic.
Section 8: 5 Strategic Imperatives for Russian Brands in 2026
Imperative 1: Monitor All 10 Providers, Not Just One
Mistake #1: companies monitor only ChatGPT because it is the most famous AI. But ChatGPT is the weakest provider for most Russian niches. Your customers use YandexGPT, Alice AI, Gemini, and Google AI Mode — those are the places to be first.
Action: launch full monitoring of all 10 providers through GEO Scout. The first data cut will show your real AI map — it will likely surprise you.
Imperative 2: Community Signals — The Unclosed Gap
According to our data, 78% of companies have a site blog, but only 23% systematically publish on Habr, vc.ru, Pikabu, or Yandex.Zen. Yet these platforms deliver disproportionately high domain citation in AI responses.
Action: build an editorial plan for external platforms. Minimum — 2 publications per month on Habr or vc.ru in your niche. This is a long-term investment: AI providers index these platforms as authoritative sources.
Imperative 3: Content Intent — Create for 9 Categories, Not 4
Informational content is the foundation, but not sufficient. Add comparative materials ("X vs Y"), recommendational ("best X for..."), and problem-solving ("how to fix the issue with..."). Each type ensures visibility in its segment of AI queries.
Action: audit your existing content by intent types. Find the "blank spots" and fill them deliberately. Taxonomy with examples is in «Content Intent Taxonomy for GEO».
Imperative 4: Optimize Entry Paths by Niche
Do not spend resources on a blog if you are a tourism service or marketplace. Optimize the home page and product pages for AI citation: Schema.org structured data, clear brand description, current data.
Action: check in GEO Scout which pages of your site AI providers already cite. That is your real AI presence map. Invest in pages that already work rather than creating new ones.
Imperative 5: Daily Monitoring — Not an Option, a Necessity
DeepSeek shifted positions for dozens of brands in a single February model update. Grok started recommending brands it had previously ignored. These changes happen within days — without daily monitoring you learn about them weeks or months later.
Action: set up daily monitoring in GEO Scout. Configure alerts for significant Mention Rate changes (±5%) and Avg Position shifts (±0.5). This lets you respond to AI provider changes within 24 hours, not quarters.
Checklist: AI Visibility for the Russian Market 2026
Use this checklist to audit your current state and plan your GEO strategy.
Monitoring and data:
- Check your brand's Mention Rate in each of the 10 providers (not just ChatGPT)
- Compare your metrics to niche benchmarks
- Identify providers where you have "zero" — they are the critical priority
- Set up daily monitoring with alerts at geoscout.pro
Content and sources:
- Create or update pages for each of the 9 Content Intent types
- Publish at least 2 articles on external platforms (Habr, vc.ru) in the next month
- Check for YouTube content about your brand (works across all providers)
- Verify your brand page on Wikipedia is accurate and current
Technical foundations:
- Implement Schema.org markup on key site pages
- Check Entry Paths: which pages do AI providers already cite for your domain?
- Optimize the home page and product pages (for e-commerce and tourism)
- Ensure robots.txt does not block AI crawlers (GPTBot, Anthropic-AI, PerplexityBot)
Provider-specific strategy:
- Create English-language content for ChatGPT (if your niche is not e-commerce)
- Strengthen presence in the Yandex ecosystem for YandexGPT/Alice AI
- Verify Claude knows your brand (its "all or nothing" pattern is the key risk)
Conclusion
2026 confirms: AI search is not the future — it is the present. 15% AI traffic already, forecast 20%+ by year-end. Google AI Mode exited beta and shows the best metrics for Russian brands. YandexGPT and Alice AI reach 88 million users — an audience that does not appear in ChatGPT statistics.
The Russian market is different from the Western one: different providers, different cited sources, different community platforms. Copying Western GEO strategies blindly is a mistake. You need to understand how AI search works specifically in Russia.
For global brands entering the Russian market, these differences are equally critical. The players that dominate AI responses here — Yandex.Market, Aviasales, Skillbox, Timeweb — built their positions through years of presence in the right sources, not through brand spending alone. Newcomers who understand the local AI landscape can accelerate this process significantly by targeting the right providers and the right platforms from day one.
State of AI Search Russia 2026 is our database on this question. We will update the report after each quarter. The next data cut — Q2 2026, July.
Full rating with daily updates, filtering by provider and niche — at geoscout.pro/ratings. Start monitoring your brand at geoscout.pro.
Частые вопросы
What is AI Search and why does it matter for the Russian market in 2026?
Which 10 AI providers does GEO Scout monitor?
Who leads AI visibility in Russian e-commerce?
How does Content Intent affect AI visibility?
Which Russian platforms do AI providers cite most often?
What is an Entry Path and why is /home more important than /blog in some niches?
How does Diversity differ between AI providers?
What is the AI traffic forecast for 2026?
Related Articles
Entry Paths in AI Search: /blog, /home, or /products — Which Page Acts as the Gateway for AI
AthenaHQ State of AI Search 2026 analysis across 8 verticals: which page type AI most commonly cites as the entry point. /blog dominates in 6 of 8 verticals, but retail and logistics play by different rules. Data, strategies, and Russian market specifics.
The Complete Content Intent Taxonomy for AI Search: 9 Categories That Change Your GEO Strategy
Why the 4 classic SEO intents fall short in GEO, and how to use all 9 Content Intent categories from AthenaHQ State of AI Search 2026. A reference guide with vertical breakdowns, format mapping, and a content audit checklist.
AI Visibility Rating: Q1 2026 Results for Russian Brands
The complete AI visibility rating across 5 niches for Q1 2026: who grew, who fell, who leads in ChatGPT, Alice, Perplexity, and other AI systems. GEO Scout monitoring data across 10 AI providers.