Each AI search platform builds its answers from almost entirely different sources. Only 1.4% of cited URLs overlap across platforms for the same query. If you are optimizing for "AI search" as one thing, you are optimizing for none of them.
You probably assume that if your page ranks well in traditional search, every AI platform will find it and cite it. The data says otherwise. Across 19,556 queries tested on five major AI platforms, the overlap in cited sources was almost nonexistent (Lee, 2026a). ChatGPT, Claude, Perplexity, Google AI Mode, and Gemini each pull from different source pools, use different retrieval architectures, and show different preferences for domain types, content formats, and even entire content categories like Reddit and YouTube.
This post breaks down exactly how each platform's citation behavior differs, what drives those differences, and what it means for your content strategy. Every claim traces back to published research or reproducible datasets.
For the full research roundup covering all major AI citation studies, see Every AI Citation Study Worth Reading in 2026. For platform-specific optimization, see our guides for ChatGPT, Perplexity, and Google AI Mode.
🔬 THE 1.4% OVERLAP PROBLEM
The single most important finding from cross-platform citation research is this: AI platforms almost never agree on which sources to cite.
Lee (2026a) tested 19,556 Google Autocomplete queries across 8 industry verticals on ChatGPT, Claude, Perplexity, and Gemini. For any given query, the Jaccard similarity of cited URLs across platforms was 0.014, meaning only 1.4% of sources appeared in more than one platform's response.
| Metric | Value | Source |
|---|---|---|
| Total queries tested | 19,556 | Lee (2026a) |
| Platforms compared | ChatGPT, Claude, Perplexity, Gemini | Lee (2026a) |
| Cross-platform URL overlap | 1.4% (Jaccard = 0.014) | Lee (2026a) |
| Within-platform consistency (ChatGPT) | 61.9% (Jaccard = 0.619) | Lee (2026a) |
| Google rank correlation with citation | rho = -0.02 to 0.11 | Lee (2026a) |
Within a single platform, consistency is much higher. ChatGPT cited the same sources 61.9% of the time when asked the same query repeatedly. But across platforms, that consistency collapsed to near-random levels.
The Bottom Line: There is no single "AI search optimization" strategy. Each platform has its own retrieval pipeline, its own source preferences, and its own architectural constraints. If you want visibility across all five major platforms, you need to understand what makes each one different.
🏗️ ARCHITECTURE DIFFERENCES: FETCHING VS. INDEXING
The reason platforms cite different sources starts with how they find content in the first place. AI search platforms fall into three distinct architectural categories, and those categories determine everything downstream.
Live Fetching Platforms (ChatGPT, Claude)
ChatGPT and Claude do not maintain their own web index. When a user asks a question that requires current information, these platforms issue search queries to an external index (Bing for ChatGPT), retrieve candidate URLs, and then fetch pages in real time using their own crawlers (ChatGPT-User and Claude-User, respectively).
This means:
- Freshness is immediate. A page updated five minutes ago can be cited.
- Server-side rendering is mandatory. These bots do not execute JavaScript. If your content loads client-side, the bot sees nothing.
- Bing indexation is the gatekeeper for ChatGPT. If Bing has not crawled your page, ChatGPT will never discover it.
- Robots.txt compliance varies. Both crawlers respect robots.txt, but with different user-agent strings.
Pre-Built Index Platforms (Perplexity)
Perplexity maintains its own web index through proactive crawling via PerplexityBot. It does not rely on a third-party search engine for URL discovery. Instead, it pre-crawls content and serves results from its index, supplemented by real-time fetches for breaking topics.
Key implications:
- Freshness has a strong bias. Perplexity's index skews 3.3x fresher than Google's for medium-velocity topics.
- Sitemap signals matter. dateModified and sitemap lastmod timestamps influence re-crawl priority.
- Crawl budget is real. Unlike live-fetching platforms, Perplexity must decide what to crawl proactively.
Google-Grounded Platforms (Google AI Mode, Gemini)
Google AI Mode and Gemini ground their responses through Google's existing search infrastructure. They inherit Google's authority signals, index depth, and ranking preferences.
Key implications:
- Traditional SEO provides the foundation. Google ranking does matter here, unlike with other AI platforms.
- No separate AI-specific crawler exists for Gemini. Content visibility depends entirely on Googlebot.
- YouTube citations are exclusive. Of 258 total YouTube citations observed across all platforms, only Google AI Mode and Perplexity cited them (Lee, 2026a). ChatGPT and Claude cited zero YouTube content.
| Platform | Architecture | URL Discovery | Real-Time Fetch? | Freshness Model |
|---|---|---|---|---|
| ChatGPT | Live fetch | Bing API | Yes | Immediate |
| Claude | Live fetch | External search | Yes | Immediate |
| Perplexity | Pre-built index | PerplexityBot | Partial | 3.3x fresher than Google |
| Google AI Mode | Google Search | Googlebot | No | Google's crawl schedule |
| Gemini | Google Search | Googlebot | No | Google's crawl schedule |
The Bottom Line: The architecture determines the rules. Optimizing for ChatGPT means ensuring Bing can find you and your pages render server-side. Optimizing for Perplexity means feeding its crawler with fresh sitemaps. Optimizing for Google AI Mode means doing good traditional SEO. These are three different playbooks, not one.
📊 CITATION VOLUME BY PLATFORM
Not all platforms cite the same number of sources per response. Some are generous with attribution. Others are sparse. This matters because the total number of citation slots available determines how competitive each platform's citation landscape is.
Based on cross-platform testing from Lee (2026a) across 19,556 queries:
| Platform | Avg. Citations Per Response | Citation Style | Notes |
|---|---|---|---|
| Perplexity | Highest | Inline numbered references | Cites the most sources per answer; designed as a "research engine" |
| Google AI Mode | High | Inline with expandable source cards | Leverages Google's deep index for broad sourcing |
| ChatGPT | Moderate | Inline bracketed links | Selective citing; often 3 to 7 sources per search-triggered response |
| Claude | Low-moderate | Inline links when search enabled | More conservative with citations; emphasizes synthesis |
| Gemini | Moderate | Inline with source chips | Grounded through Google Search results |
The practical implication: getting cited by Perplexity is statistically easier because it cites more sources per response. Getting cited by Claude is harder because it cites fewer. But Perplexity's larger citation volume also means the competition for each slot is different.
For a detailed head-to-head comparison of these platforms and how to optimize for each, see ChatGPT vs Perplexity vs Gemini: Which AI Search Engine Should You Optimize For?.
🌐 DOMAIN TYPE PREFERENCES BY PLATFORM
Each AI platform shows measurable preferences for different types of source domains. These preferences reflect both the platform's architecture and its design philosophy.
From the cross-platform analysis in Lee (2026a), domain-level alignment with Google was 4 to 7 times stronger than URL-level alignment. Platforms tend to draw from the same top domains as Google (Wikipedia, major publishers, .gov/.edu sites) but select entirely different pages within those domains.
| Domain Type | ChatGPT | Claude | Perplexity | Google AI Mode | Gemini |
|---|---|---|---|---|---|
| Wikipedia/encyclopedic | High | High | Moderate | High | High |
| .gov/.edu authoritative | Moderate | Moderate | Moderate | High | High |
| News/media publishers | Moderate | Low | High | High | Moderate |
| Review aggregators | Moderate | Low | High | Moderate | Moderate |
| Brand/company sites | Low | Low | Moderate | Moderate | Moderate |
| 0% API / 17% web | 0% both | 0% API / 20% web | 44% web | Low | |
| YouTube | 0% | 0% | Present | Present | Low |
The domain preferences reflect each platform's retrieval strategy. Perplexity and Google AI Mode, which maintain their own indexes, have broader domain coverage including news publishers and review sites. ChatGPT and Claude, which fetch on demand, tend toward well-known authoritative domains that surface reliably through Bing.
The Bottom Line: Your domain type influences which platforms are most likely to cite you. News publishers have an advantage on Perplexity and Google AI Mode. Authoritative reference content performs well on ChatGPT and Claude. Brand sites struggle across all platforms unless the query specifically names the brand.
👻 THE REDDIT DIVERGENCE: 0% API VS. 17-44% WEB UI
One of the most striking findings in AI citation research is how differently platforms treat Reddit content depending on the access channel.
Lee (2026b) tested identical brand recommendation queries through both API access (programmatic) and web UI access (browser-based) across all major platforms. The divergence was dramatic:
| Platform | Reddit Citations (API) | Reddit Citations (Web UI) | Gap |
|---|---|---|---|
| Google AI Mode | 0% | 44% | 44 percentage points |
| Perplexity | 0% | 20% | 20 percentage points |
| ChatGPT | 0% | 17% | 17 percentage points |
| Claude | 0% | 0% | None |
Through API access, Reddit was cited exactly zero percent of the time across every platform. Through the web UI, Reddit appeared in up to 44% of Google AI Mode responses. Yet the actual brand recommendations were similar across both channels, suggesting that Reddit's influence operates through training data absorption rather than real-time citation.
Why This Happens
The access-channel split reveals three distinct pathways for Reddit influence:
Training data pathway. Reddit content has been absorbed into LLM training data. Models draw on internalized Reddit sentiment without citing a source. Lee (2026b) found a Spearman correlation of rho = 0.554 between Reddit brand mention frequency and AI recommendation probability.
Web UI citation pathway. When search is enabled in the browser interface, platforms find and cite live Reddit threads that match already-formed opinions. Aggregate web UI Reddit citation rate: 27%.
API citation pathway. Programmatic access suppresses Reddit citations categorically. Zero percent across all platforms.
The implication for content creators: Reddit is a "shadow corpus." It shapes AI outputs even when it is never cited. If your brand has negative sentiment on Reddit, that sentiment is likely influencing AI recommendations regardless of what your website says.
For validation queries specifically (those seeking opinions and comparisons), Reddit citation rates spiked even higher: 71% on Google AI Mode and 46% on Perplexity in the web UI (Lee, 2026b). Intent drives Reddit visibility.
For the full analysis of Reddit's invisible influence on AI, see our Query Intent and AI Citation research.
🎬 YOUTUBE CITATIONS: THE PLATFORM-EXCLUSIVE CHANNEL
YouTube represents another clear example of platform-specific citation behavior. Across all queries tested in Lee (2026a), a total of 258 YouTube citations were observed. But they were not distributed evenly.
| Platform | YouTube Citations | Notes |
|---|---|---|
| Google AI Mode | Present (majority share) | Leverages Google's ownership of YouTube |
| Perplexity | Present (smaller share) | Includes video results in its index |
| ChatGPT | 0 | Does not cite YouTube content |
| Claude | 0 | Does not cite YouTube content |
| Gemini | Minimal | Despite Google ownership, lower than AI Mode |
The exclusivity makes sense architecturally. Google AI Mode has native access to YouTube's content graph. Perplexity's proactive crawler indexes video metadata. ChatGPT and Claude, which rely on live page fetching, do not process video content during real-time retrieval.
The Bottom Line: If your content strategy includes video, your YouTube presence will only generate AI citations on Google AI Mode and Perplexity. For ChatGPT and Claude visibility, video content needs a companion text page (transcript, blog post, or summary) hosted on your own domain.
🎯 INTENT-DRIVEN CITATION PATTERNS
The strongest predictor of which sources get cited is not domain authority, backlinks, or page speed. It is whether the content matches the query's intent.
Lee (2026a) classified all 19,556 queries into five intent categories and found that intent distributions varied significantly by vertical (chi-squared(28) = 5,195, p < .001, Cramer's V = 0.258). Each intent type surfaces a different category of sources:
| Intent Type | Query Share | Typical Citation Sources | Platform Variation |
|---|---|---|---|
| Informational | 61.3% | Wikipedia, .gov/.edu, tutorials | Consistent across platforms |
| Discovery | 31.2% | Review aggregators, YouTube, listicles | Perplexity cites broadest range |
| Validation | 3.2% | Brand sites, Reddit (web UI only) | Reddit rates vary 0-71% by platform |
| Comparison | 2.3% | Publisher/media, review sites | ChatGPT and Perplexity diverge most |
| Review-seeking | 2.0% | YouTube, TechRadar/PCMag, Reddit | Google AI Mode dominates YouTube |
A critical finding from the regression analysis: adding intent features to a page-level prediction model provided zero additional predictive power (likelihood ratio p = .78). This means intent acts as a filter, not a feature. It determines which content pool is eligible. Page-level features (internal links, content-to-HTML ratio, schema markup) then determine which page wins within that pool.
The GEO framework from Aggarwal et al. (2024) supports this finding from the optimization side. Their research showed that different optimization strategies work in different domains, with "citing sources" and "adding statistics" producing up to 40% visibility improvement in factual domains but minimal impact in opinion-based ones. The optimization that works depends on the intent the content targets.
For a practical framework on matching your content to query intent, see our ChatGPT SEO Optimization Guide or run a free check with our AI Visibility Quick Check.
📈 WHAT THIS MEANS FOR YOUR STRATEGY
The cross-platform comparison data points to a clear strategic framework:
Step 1: Accept Platform Fragmentation
With 1.4% citation overlap, there is no shortcut. You need to understand each platform's retrieval method and optimize accordingly. A page that gets cited by Perplexity may be invisible to ChatGPT, and vice versa.
Step 2: Match Architecture to Tactics
| If You Want Citations From... | Prioritize... |
|---|---|
| ChatGPT | Bing indexation, server-side rendering, clean HTML |
| Claude | Server-side rendering, high content-to-HTML ratio |
| Perplexity | Fresh sitemaps, dateModified signals, proactive crawl access |
| Google AI Mode | Traditional Google SEO, YouTube presence |
| Gemini | Google Search visibility, structured data |
Step 3: Front-Load for All Platforms
One finding that holds across every platform: 44.2% of citations reference content from the first 30% of the page (Sellm, 2025). Regardless of which platform you target, put your most citation-worthy information at the top.
Step 4: Use the Right Schema Types
Lee (2026a) found that schema type matters more than schema presence. Product schema (OR = 3.09), Review schema (OR = 2.24), and FAQPage schema (OR = 1.39) all increased citation probability. Article schema actually decreased it (OR = 0.76). Apply schema strategically, not generically.
Step 5: Monitor Reddit as an Upstream Signal
Even at 0% API citation rate, Reddit influences AI recommendations through training data. Track your brand's Reddit sentiment as an input to AI outputs, not just as a social media channel.
❓ FREQUENTLY ASKED QUESTIONS
Which AI platform cites the most sources per response?
Perplexity consistently cites the highest number of sources per answer, followed by Google AI Mode. Perplexity was designed as a research-oriented engine with inline numbered references, giving it the broadest citation footprint. ChatGPT and Claude are more selective, typically citing 3 to 7 sources when web search is triggered (Lee, 2026a).
Why do AI platforms cite completely different sources for the same query?
Three reasons: different retrieval architectures (live fetch vs. pre-built index vs. Google Search), different URL discovery methods (Bing vs. PerplexityBot vs. Googlebot), and different content evaluation criteria. The 1.4% overlap (Lee, 2026a) is a direct consequence of these architectural differences. Each platform effectively searches a different version of the web.
Does Google ranking help with AI citations on non-Google platforms?
Minimally. Lee (2026a) found the correlation between Google rank and AI citation ranged from rho = -0.02 to 0.11 across ChatGPT, Claude, and Perplexity. Domain-level alignment was stronger (28.7% to 49.6%), meaning platforms draw from similar top domains but select different pages. For Google AI Mode and Gemini, Google ranking matters significantly more because those platforms are grounded in Google Search.
How does Reddit influence AI answers if it is never cited through the API?
Reddit operates as a "shadow corpus" (Lee, 2026b). Its content was absorbed into LLM training data, creating a Spearman correlation of rho = 0.554 between Reddit brand consensus and AI brand recommendations. Through the API, this influence is invisible (0% citation rate). Through web UIs, Reddit is cited 17% to 44% of the time, depending on the platform and query type.
Can YouTube content help me get cited by AI platforms?
Only on Google AI Mode and Perplexity. Of 258 total YouTube citations observed (Lee, 2026a), ChatGPT and Claude cited zero YouTube content. If you invest in video, create companion text content on your own domain to capture citations from fetching-based platforms.
📚 REFERENCES
- Aggarwal, P., Murahari, V., Rajpurohit, T., Kalyan, A., Narasimhan, K., & Deshpande, A. (2024). "GEO: Generative Engine Optimization." KDD 2024. DOI
- Lee, A. (2026a). "Query Intent, Not Google Rank: What Best Predicts AI Citation Behavior." Preprint v5. DOI
- Lee, A. (2026b). "Reddit Doesn't Get Cited (Through the API): Training Data Influence, Access-Channel Divergence, and the Shadow Corpus in AI Brand Recommendations." Preprint. DOI
- Sellm (2025). "ChatGPT Citation Analysis." Industry report (400K+ pages analyzed).