The rapid integration of AI chatbots into consumer search behavior has spawned a cottage industry of Generative Engine Optimization (GEO) advice, much of it built on untested assumptions about how AI platforms select sources for citation. Industry practitioners widely assert that Google ranking determines AI visibility, that community-consensus platforms like Reddit confer citation advantages, and that AI recommendations are too inconsistent to warrant optimization efforts. We tested these claims empirically across four major AI platforms — ChatGPT, Claude, Perplexity, and Gemini — using a multi-study design that combined large-scale query intent classification (*n* = 19,556 queries across 8 verticals), Google rank cross-referencing (120 queries via API, plus 100 queries via web UI against both Google and Bing Top-3 results), server-side fetch verification via Vercel middleware logging, and page-level technical analysis of 479 cited and non-cited pages. Our results refine all three prevailing claims. First, query intent shaped citation source-type distributions across verticals (χ²(28) = 5,195, *p* < .001, Cramér's *V* = 0.258), but at the individual-page level Google rank dominated all other predictors: a logistic regression with log(Google position) alone achieved cross-validated AUC = 0.802 and McFadden *R*² = 0.203, far exceeding intent-only (AUC = 0.462) and page-feature-only (AUC = 0.594) baselines. A larger replication on 94,599 citation events across 1,998 queries (Lee, 2026c — the *SEO Floor* study) found a steep monotonic position gradient: URLs at Google position 1 were cited by ≥1 AI platform 54% of the time, dropping to ≈2% at position 100, with every platform showing a 13–22× citation-rate ratio between Top-3 and positions 31–100. Google rank therefore predicts citation probability strongly, even though the cited URL frequently does not appear in Google's literal-query Top-3 (ChatGPT 7.8%, Perplexity 29.7% — replicated and confirmed under Gemini-reformulated queries: 7.8% / 35.4%). Domain-level analysis (Study 2b) revealed substantially higher alignment than URL-level (28.7–49.6% across four platforms), indicating that AI platforms draw from Google's top-ranked domains while frequently selecting different specific pages. A dual search engine comparison showed that all platforms aligned 4–7× more strongly with Google than with Bing organic results, even platforms reportedly using Bing as their search backend. Reddit — despite occupying 38.3% of Google Top-3 positions in our API sample — received exactly zero AI citations via API (binomial *p* = 3.43 × 10⁻²³ for Perplexity), though web UI responses from the same platforms cited Reddit at rates of 8.9–15.6% of total citations. Third, AI brand recommendations showed substantial within-platform consistency (ChatGPT mean Jaccard = 0.619, 95% CI 0.537, 0.701), though cross-platform agreement was near-random (all-four-platform Jaccard = 0.036). We further discovered a previously unreported architectural divide: ChatGPT and Claude perform live page fetches during conversations, while Perplexity and Gemini rely exclusively on pre-built search indices — with divergent robots.txt compliance behavior between the fetching platforms. These findings suggest that effective GEO strategy requires intent-aware, platform-specific optimization rather than the one-size-fits-all approach currently advocated by industry practitioners.
Anthony Lee (Sun,) studied this question.