GEO Lab #04 · S4-OPENAIAPI-Q2-R2 · Original response
This page presents one selected response from GEO Lab #04. It adds a smaller reading and citation entry point; it does not replace the full archive or collect a new answer.
The OpenAI arm uses the Responses API with web search and does not represent consumer ChatGPT Search. Grok records identify their actual consumer-interface mode and available model label.
Recorded conditions
Original prompt
What credible experiments have measured whether GEO tactics increase AI brand mentions or recommendations? Include sources when available.
The agreed prompt is reproduced exactly. Any recorded difference in the actual submitted text or search UI is retained in the raw record and full appendix.
Original answer
## Short answer **Yes—but the strongest evidence comes from controlled or adversarial experiments, not ordinary brand-marketing campaigns.** These studies show that changing source content can increase a product’s inclusion, ranking, or recommendation by an LLM. Evidence that conventional GEO programs reliably increase **real-world, non-branded mentions across live ChatGPT, Gemini, or Perplexity** remains limited and mostly consists of uncontrolled before/after case studies. ### Best available experiments | Experiment | Intervention and outcome | Credibility and limitations | |---|---|---| | **GEO: Generative Engine Optimization — KDD 2024** | Across a 10,000-query benchmark, adding relevant statistics, credible quotations, and citations increased source visibility roughly 30–40% on some metrics. In a 200-query Perplexity test, the best methods improved position-adjusted visibility by 22% and subjective visibility by 37%. Keyword stuffing generally performed poorly. ([arxiv.org](https://arxiv.org/abs/2311.09735)) | **Strong, peer-reviewed controlled evidence**, but it measured how much an already-retrieved source influenced an answer—not spontaneous brand mentions. The Perplexity experiment uploaded the source documents directly, so it did not test whether normal web publishing caused retrieval. | | **Manipulating Large Language Models to Increase Product Visibility — Kumar & Lakkaraju, 2024** | Researchers inserted optimized “strategic text sequences” into descriptions of fictitious coffee machines. The intervention significantly increased the likelihood that low-ranked target products became the LLM’s top recommendation. Code was released. ([arxiv.org](https://arxiv.org/abs/2404.07981)) | **Strong causal design**, but the tactic resembles adversarial optimization rather than normal content marketing, and the catalog was synthetic rather than the live web. | | **Adversarial Search Engine Optimization for LLMs — ICLR 2025** | Researchers published fictitious cameras and tested them against established products such as Nikon and Fujifilm. A preference-manipulation intervention nearly doubled the fictitious product’s recommendation rate, making it slightly more likely to be recommended than the established products. ([openreview.net](https://openreview.net/pdf?id=hkdqxN3c7t)) | **Peer-reviewed and closer to a deployed search setting.** However, prompts explicitly directed the engine to search the researchers’ domain, which may exaggerate real-world effectiveness. | | **SafeGEO, 2026** | Across 600 recommendation cases and 22 content-manipulation variants, seller-controlled rewrites increased the rate at which a flawed target product appeared in the top three by as much as **83.2 percentage points**. Some rewrites also produced recommendations that violated users’ stated constraints. ([arxiv.org](https://arxiv.org/abs/2606.28356)) | **Large, carefully controlled benchmark**, with truthful-rewrite controls. It demonstrates that source wording can cause recommendation uplift—but primarily as a safety or manipulation result, not proof that ethical GEO works equally well on live consumer engines. | | **Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics, 2026** | Tests across GPT-4o-mini, Claude Sonnet, and Gemini found that authority-style marketing language could overcome an incumbent brand advantage. In the paper’s setup, the language had an effect equivalent to roughly a 0.17-star rating advantage. ([arxiv.org](https://arxiv.org/abs/2606.17443)) | **Useful multi-model controlled evidence**, but currently a preprint. Some treatments included fabricated clinical-evidence claims, making the results more relevant to manipulation risk than legitimate GEO practice. | | **SE Ranking fictional-brand field experiment, 2026** | A fictional brand was established on a new site and discussed across 11 additional domains using guides, comparisons, reviews, “best of” lists and other formats. Researchers tracked 825 prompts and 15,835 answers across ChatGPT, Google AI Overviews, AI Mode, Perplexity and Gemini. The brand gained rapid visibility for branded and uniquely answerable queries, but 96% of its total visibility came from branded searches; it made little progress on broad non-branded topics. ([searchengineland.com](https://searchengineland.com/fake-brand-ai-search-experiment-475947)) | **One of the more relevant live-web interventions**, but not randomized: multiple tactics and domains changed simultaneously, there was no untreated comparison brand, and the first results covered only about one month. | | **Training-time brand imprint study, ktau, 2026** | Researchers added 250 engineered documents to a 30B model’s training corpus. A previously invisible fictional brand reached a 42.1% recommendation rate and 44.3% mention rate on broad held-out queries. ([ktau.ai](https://www.ktau.ai/research/training-time-brand-imprint-study)) | **Interesting controlled result**, but lower evidentiary weight: company-authored, not peer-reviewed, and dependent on direct training-corpus access—a channel ordinary brands cannot control reliably. | ## Important counterevidence about popular GEO tactics A large 2026 study analyzed **158,099 commercial-query answers and 468,037 citations** over two 90-day periods. Its within-website comparisons found no significant citation benefit from FAQ blocks, schema markup, summary boxes, author boxes, or page speed. Apparent benefits largely disappeared after controlling for the fact that stronger websites were more likely to use those features. ([arxiv.org](https://arxiv.org/abs/2610.15033)) Instead, the important factors were: - Semantic alignment between page titles/text and the subqueries generated by the AI engine. - Whether the engine retrieved the page at all. - Retrieval rank. - The model’s pre-existing knowledge of the brand. For ChatGPT, retrieval variables explained about **32%** of within-site citation variation, compared with **2.8%** for measured page features. The study also found that 46% of ChatGPT brand mentions lacked a supporting cited page, indicating that model memory—not merely current web optimization—affects mentions. ([arxiv.org](https://arxiv.org/abs/2610.15033)) This was primarily observational rather than a publishing intervention, but its within-site design makes it substantially more informative than simple correlation studies. ## Commercial before/after case studies These provide field evidence, but should not be treated as controlled experiments: - **Gradial:** reported a rise from roughly 2–3% to 26% mention rate after a site redesign informed by GEO analysis. The company explicitly says the redesign “did the heavy lifting,” and provides no control group, repeated-run counts, or tactic-level isolation. ([gradial.com](https://www.gradial.com/blog/gradial-improved-ai-mentions-from-2-to-26-percent)) - **Smoking Gun PR:** reported AI share of voice rising from 3.2% to 9.1% and tracked mentions rising from 274 to 1,424 after a four-month earned-media program. The client was unnamed and the public case study does not disclose the prompt panel, engines, run frequency, or control category. ([smokinggun.agency](https://smokinggun.agency/our-work/geo-casestudy/)) - **Growthino:** reported its first non-branded citation after six weeks and six ChatGPT-attributed sessions during a two-day analytics window by week ten. It is transparent about the small sample, but one citation and a handful of sessions cannot establish causality. ([growthino.com](https://growthino.com/case-studies/from-zero-to-cited-by-ai-in-6-weeks)) ## Bottom line The evidence supports three relatively defensible conclusions: 1. **Source content can causally alter AI recommendations** when the source is placed in the model’s context or retrieved by the system. This is well demonstrated by controlled experiments. 2. **Retrievability and query-level semantic relevance appear more important than cosmetic GEO additions** such as FAQ schema, summary boxes, or merely adopting an “authoritative” tone. 3. **There is not yet strong public evidence that a standard ethical GEO campaign consistently increases non-branded recommendations across live AI engines.** The live-web evidence is mostly multi-intervention, pre/post, vendor-reported, and vulnerable to model updates, prompt sampling, personalization, and normal run-to-run variation. A convincing commercial test would need preregistered prompts, repeated runs, engine-specific reporting, concurrent untreated pages or brands, staggered interventions, and separate measurement of **citations, mentions, recommendation position, and final recommendation rate**.
Recorded sources
These are the source roles recorded in the public dataset. A retrieval or source-panel item is not automatically an answer-body citation. A URL supplied in the prompt and repeated by an answer does not prove content access. No destination has been repaired or independently validated here.
Recorded answer-body links (12)
- GEO: Generative Engine OptimizationAPI body url_citation annotation · arxiv.org
- Manipulating Large Language Models to Increase Product VisibilityAPI body url_citation annotation · arxiv.org
- Adversarial Search Engine Optimization for Large Language ModelsAPI body url_citation annotation · openreview.net
- SafeGEO: Understanding Generative Engine Optimization Risks in Recommendation AgentsAPI body url_citation annotation · arxiv.org
- Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation SystemsAPI body url_citation annotation · arxiv.org
- Can a fake brand win in AI search? New experiment says yesAPI body url_citation annotation · searchengineland.com
- From Invisible to Recommended: A Controlled 30B-Parameter Study of Training-Time Brand Imprint - ktauAPI body url_citation annotation · www.ktau.ai
- What Drives AI Citations? Content, Retrieval and Model Knowledge in Generative Engine Optimization (GEO)API body url_citation annotation · arxiv.org
- What Drives AI Citations? Content, Retrieval and Model Knowledge in Generative Engine Optimization (GEO)API body url_citation annotation · arxiv.org
- We Took Our GEO Mention Rate from 2% to 26% in Four Weeks. Here's What Actually Moved the Number. | GradialAPI body url_citation annotation · www.gradial.com
- Moving the dial on AI discovery: A GEO case study in LLM visibility – Smoking Gun PRAPI body url_citation annotation · smokinggun.agency
- From Zero to Cited by AI: A 10-Week GEO Experiment | GrowthinoAPI body url_citation annotation · growthino.com
Explicit plain-text body URLs (11)
- https://arxiv.org/abs/2311.09735Explicit plain-text address, not an inline citation · arxiv.org
- https://arxiv.org/abs/2404.07981Explicit plain-text address, not an inline citation · arxiv.org
- https://openreview.net/pdf?id=hkdqxN3c7tExplicit plain-text address, not an inline citation · openreview.net
- https://arxiv.org/abs/2606.28356Explicit plain-text address, not an inline citation · arxiv.org
- https://arxiv.org/abs/2606.17443Explicit plain-text address, not an inline citation · arxiv.org
- https://searchengineland.com/fake-brand-ai-search-experiment-475947Explicit plain-text address, not an inline citation · searchengineland.com
- https://www.ktau.ai/research/training-time-brand-imprint-studyExplicit plain-text address, not an inline citation · www.ktau.ai
- https://arxiv.org/abs/2610.15033Explicit plain-text address, not an inline citation · arxiv.org
- https://www.gradial.com/blog/gradial-improved-ai-mentions-from-2-to-26-percentExplicit plain-text address, not an inline citation · www.gradial.com
- https://smokinggun.agency/our-work/geo-casestudy/Explicit plain-text address, not an inline citation · smokinggun.agency
- https://growthino.com/case-studies/from-zero-to-cited-by-ai-in-6-weeksExplicit plain-text address, not an inline citation · growthino.com
Retrieval-only / source-panel inventory (105)
- https://www.axios.com/2026/10/28/ai-citation-source-georetrieval source, not automatically a body citation · www.axios.com
- https://www.techradar.com/pro/hubspot-aeo-reviewretrieval source, not automatically a body citation · www.techradar.com
- https://www.measure.co/features/ai-visibility-trackingretrieval source, not automatically a body citation · www.measure.co
- https://presenc.ai/research/brand-visibility-in-ai-studyretrieval source, not automatically a body citation · presenc.ai
- https://smokinggun.agency/our-work/geo-casestudy/retrieval source, not automatically a body citation · smokinggun.agency
- https://www.gradial.com/blog/gradial-improved-ai-mentions-from-2-to-26-percentretrieval source, not automatically a body citation · www.gradial.com
- https://scrunch.com/blog/prompt-to-purchase-pipeline-how-ai-influences-buyer-behavior?content_language=English&facet3=pdfretrieval source, not automatically a body citation · scrunch.com
- https://www.traillabs.ai/blog/geo-at-scale-brand-visibilityretrieval source, not automatically a body citation · www.traillabs.ai
- https://growthino.com/case-studies/from-zero-to-cited-by-ai-in-6-weeksretrieval source, not automatically a body citation · growthino.com
- https://arxiv.org/abs/2606.28356retrieval source, not automatically a body citation · arxiv.org
- https://searchengineland.com/fake-brand-ai-search-experiment-475947retrieval source, not automatically a body citation · searchengineland.com
- https://stanislav-peev.com/research/ai-visibility-tracking-reliability/retrieval source, not automatically a body citation · stanislav-peev.com
- https://github.com/princy2310/ai-visibility-auditretrieval source, not automatically a body citation · github.com
- https://humanswith.ai/cases/humanswith-ai-aeo-geo-2/retrieval source, not automatically a body citation · humanswith.ai
- https://getrefine.ai/blog/ai-visibility-benchmark-17691-runsretrieval source, not automatically a body citation · getrefine.ai
- https://arxiv.org/abs/2610.24081retrieval source, not automatically a body citation · arxiv.org
- https://arxiv.org/abs/2509.08919retrieval source, not automatically a body citation · arxiv.org
- https://arxiv.org/abs/2311.09735retrieval source, not automatically a body citation · arxiv.org
- https://geometriqs.com/download/GenAI%20Global80%20Report%20-%20Geometriqs%20Nov%202025.pdfretrieval source, not automatically a body citation · geometriqs.com
- https://aigeo.games/media/research/generative-engine-optimization-how-to-dominate-ai-search.pdfretrieval source, not automatically a body citation · aigeo.games
- https://www.reddit.com/r/GEO_optimization/comments/1ulhqto/i_tested_whether_ai_search_is_one_ranking_or/retrieval source, not automatically a body citation · www.reddit.com
- https://www.reddit.com/r/GEO_optimization/comments/1uis2co/some_findings_from_working_on_branded_ai/retrieval source, not automatically a body citation · www.reddit.com
- https://www.tandfonline.com/doi/pdf/10.1080/13527266.2025.2525913retrieval source, not automatically a body citation · www.tandfonline.com
- https://thegeolab.net/wp-content/uploads/ebooks/GEO-Experiments-Templates.pdfretrieval source, not automatically a body citation · thegeolab.net
- https://cdn0.scrvt.com/6d226c03f32a8ea4e8ae8e5cbe6c6e2c/30a619ef951f44e3/48019f031474/Generative-Engine-Optimization-whitepaper.pdfretrieval source, not automatically a body citation · cdn0.scrvt.com
- https://journals.sagepub.com/doi/pdf/10.1177/07439156241297973retrieval source, not automatically a body citation · journals.sagepub.com
- https://www.reddit.com/r/GenerativeSEOstrategy/comments/1ulhszj/is_ai_search_one_ranking_or_three_different_ones/retrieval source, not automatically a body citation · www.reddit.com
- https://www.reddit.com/r/SEO_LLM/comments/1v9tehz/anyone_here_successfully_improved_their/retrieval source, not automatically a body citation · www.reddit.com
- https://www.reddit.com/r/b2bmarketing/comments/1sg77pd/we_tracked_147_prompts_across_chatgpt_gemini/retrieval source, not automatically a body citation · www.reddit.com
- https://www.reddit.com/r/GEO_optimization/comments/1uqjxvv/we_tested_60_brands_across_4_ai_conversation/retrieval source, not automatically a body citation · www.reddit.com
- https://www.reddit.com/r/AISearchLab/comments/1v6w520/i_tested_whether_ai_visibility_tools_are_actually/retrieval source, not automatically a body citation · www.reddit.com
- https://www.reddit.com/r/AISEOInsider/comments/1tur04g/i_ran_a_small_geo_experiment_to_see_what_actually/retrieval source, not automatically a body citation · www.reddit.com
- https://www.reddit.com/r/GenEngineOptimization/comments/1ry1cqs/most_geo_tools_measure_visibility_not_outcome_6/retrieval source, not automatically a body citation · www.reddit.com
- https://www.reddit.com/r/GEO_optimization/comments/1tndp3q/spent_a_day_running_the_same_brand_queries/retrieval source, not automatically a body citation · www.reddit.com
- https://www.reddit.com/r/SEO/comments/1rj2mid/i_checked_if_chatgpt_recommends_my_brand_the/retrieval source, not automatically a body citation · www.reddit.com
- https://www.reddit.com/r/SideProject/comments/1rj2or4/i_checked_if_chatgpt_recommends_my_brand_the/retrieval source, not automatically a body citation · www.reddit.com
- https://de.wikipedia.org/wiki/Answer_Engine_Optimizationretrieval source, not automatically a body citation · de.wikipedia.org
- https://en.wikipedia.org/wiki/Advertising_researchretrieval source, not automatically a body citation · en.wikipedia.org
- https://pt.wikipedia.org/wiki/Generative_engine_optimizationretrieval source, not automatically a body citation · pt.wikipedia.org
- https://en.wikipedia.org/wiki/Algorithmic_amplificationretrieval source, not automatically a body citation · en.wikipedia.org
- https://fr.wikipedia.org/wiki/Optimisation_pour_les_moteurs_g%C3%A9n%C3%A9ratifsretrieval source, not automatically a body citation · fr.wikipedia.org
- https://www.techradar.com/pro/surfer-ai-tracker-reviewretrieval source, not automatically a body citation · www.techradar.com
- https://www.theatlantic.com/technology/2026/06/google-search-ai-optimization/687495/retrieval source, not automatically a body citation · www.theatlantic.com
- https://www.axios.com/2026/08/20/chatgpt-reddit-citations-geo-strategyretrieval source, not automatically a body citation · www.axios.com
- https://www.axios.com/newsletters/axios-communicators-c880e9e0-9bdf-11f1-9602-2f20038b64b1retrieval source, not automatically a body citation · www.axios.com
- https://en.wikipedia.org/wiki/Search_engine_optimizationretrieval source, not automatically a body citation · en.wikipedia.org
- https://www.theatlantic.com/technology/archive/2024/06/chatgpt-citations-rag/678796/retrieval source, not automatically a body citation · www.theatlantic.com
- https://www.techradar.com/pro/i-tested-the-7-best-aeo-toolsretrieval source, not automatically a body citation · www.techradar.com
- https://www.techradar.com/pro/hubspot-aeo-reviewretrieval source, not automatically a body citation · www.techradar.com
- https://www.ktau.ai/research/training-time-brand-imprint-studyretrieval source, not automatically a body citation · www.ktau.ai
- https://arxiv.org/abs/2610.15033retrieval source, not automatically a body citation · arxiv.org
- https://arxiv.org/abs/2606.28356retrieval source, not automatically a body citation · arxiv.org
- https://arxiv.org/abs/2606.17443retrieval source, not automatically a body citation · arxiv.org
- https://arxiv.org/abs/2404.07981retrieval source, not automatically a body citation · arxiv.org
- https://www.clearscope.io/blog/clearscope-aeo-experimentretrieval source, not automatically a body citation · www.clearscope.io
- https://ai-cdn.scrunchai.com/blog/prompt-to-purchase-pipeline-how-ai-influences-buyer-behavior/retrieval source, not automatically a body citation · ai-cdn.scrunchai.com
- https://users.ece.cmu.edu/~lbauer/papers/2025/chi2025-LLM-brands-bias-noacm.pdfretrieval source, not automatically a body citation · users.ece.cmu.edu
- https://www.alphaxiv.org/abs/2604.04263retrieval source, not automatically a body citation · www.alphaxiv.org
- https://www.adlift.com/blog/logged-in-or-logged-out-do-ai-platforms-choose-different-brands-to-recommend/retrieval source, not automatically a body citation · www.adlift.com
- https://hdsr.mitpress.mit.edu/pub/2zvzqo4k/download/pdfretrieval source, not automatically a body citation · hdsr.mitpress.mit.edu
- https://onlinelibrary.wiley.com/doi/10.1002/mar.21706retrieval source, not automatically a body citation · onlinelibrary.wiley.com
- https://papers.ssrn.com/sol3/Delivery.cfm/7208681.pdf?abstractid=7208681&mirid=1&type=2retrieval source, not automatically a body citation · papers.ssrn.com
- https://auracite.de/content/research/state-of-ai-visibility-2026.htmlretrieval source, not automatically a body citation · auracite.de
- https://www.amazon.science/publications/llms-for-customized-marketing-content-generation-and-evaluation-at-scaleretrieval source, not automatically a body citation · www.amazon.science
- https://www.aimodels.fyi/papers/arxiv/one-polluted-page-is-enough-evaluating-webretrieval source, not automatically a body citation · www.aimodels.fyi
- https://huggingface.co/papers/2404.07981retrieval source, not automatically a body citation · huggingface.co
- https://www.mygosh.ai/the-authority-gapretrieval source, not automatically a body citation · www.mygosh.ai
- https://openreview.net/pdf?id=hkdqxN3c7tretrieval source, not automatically a body citation · openreview.net
- https://www.econstor.eu/bitstream/10419/336742/1/194960134X.pdfretrieval source, not automatically a body citation · www.econstor.eu
- https://www.nim.org/fileadmin/PUBLIC/12_NIM_MIR_Issues/MIR_Generative_KI/MIR_Generative_AI/2024_nim_mir_Generative-AI_en.pdfretrieval source, not automatically a body citation · www.nim.org
- https://www.reddit.com/r/AISearchLab/comments/1ufxtmd/i_tested_15_ai_searches_about_one_brand_even/retrieval source, not automatically a body citation · www.reddit.com
- https://en.wikipedia.org/wiki/Search_engine_manipulation_effectretrieval source, not automatically a body citation · en.wikipedia.org
- https://en.wikipedia.org/wiki/Algorithmic_amplificationretrieval source, not automatically a body citation · en.wikipedia.org
- https://en.wikipedia.org/wiki/Recommender_systemretrieval source, not automatically a body citation · en.wikipedia.org
- https://www.theatlantic.com/technology/2026/06/google-search-ai-optimization/687495/retrieval source, not automatically a body citation · www.theatlantic.com
- https://www.axios.com/2026/06/25/ai-search-collapse-geo-seoretrieval source, not automatically a body citation · www.axios.com
- https://www.techradar.com/pro/why-ai-visibility-now-demands-paid-and-organic-geo-optimizationretrieval source, not automatically a body citation · www.techradar.com
- https://pt.wikipedia.org/wiki/Generative_engine_optimizationretrieval source, not automatically a body citation · pt.wikipedia.org
- https://en.wikipedia.org/wiki/Personalized_searchretrieval source, not automatically a body citation · en.wikipedia.org
- https://en.wikipedia.org/wiki/Generative_engine_optimizationretrieval source, not automatically a body citation · en.wikipedia.org
- https://seranking.com/api/project/airt-competitors/retrieval source, not automatically a body citation · seranking.com
- https://seranking.com/blog/ai-content-experiment/retrieval source, not automatically a body citation · seranking.com
- https://seranking.com/blog/search-intent/retrieval source, not automatically a body citation · seranking.com
- https://visible.seranking.com/blog/ai-overviews-branded-searches/retrieval source, not automatically a body citation · visible.seranking.com
- https://help.seranking.com/hc/en-us/articles/26158575790620-Know-how-your-brand-appears-in-AI-search-across-all-engines-at-scaleretrieval source, not automatically a body citation · help.seranking.com
- https://help.seranking.com/hc/en-us/articles/22449302486428-AI-Search-in-Competitive-Researchretrieval source, not automatically a body citation · help.seranking.com
- https://seranking.com/blog/how-to-optimize-for-ai-mode/retrieval source, not automatically a body citation · seranking.com
- https://visible.seranking.com/blog/choose-an-ai-visibility-platform/retrieval source, not automatically a body citation · visible.seranking.com
- https://seranking.com/blog/ai-mode-seo-impact/retrieval source, not automatically a body citation · seranking.com
- https://visible.seranking.com/blog/studies-experiments-category/retrieval source, not automatically a body citation · visible.seranking.com
- https://seranking.com/blog/brand-evolution/retrieval source, not automatically a body citation · seranking.com
- https://visible.seranking.com/chatgpt-brand-monitoring/retrieval source, not automatically a body citation · visible.seranking.com
- https://seranking.com/blog/wp-content/uploads/2020/12/SE_Ranking_Sample_SEO_Report.pdfretrieval source, not automatically a body citation · seranking.com
- https://seranking.com/de/blog/wp-content/uploads/sites/40/2022/12/Ein-fast-perfekter-SEO-Bericht.pdfretrieval source, not automatically a body citation · seranking.com
- https://arxiv.org/abs/2311.09735retrieval source, not automatically a body citation · arxiv.org
- https://arxiv.org/abs/2404.07981retrieval source, not automatically a body citation · arxiv.org
- https://openreview.net/pdf?id=hkdqxN3c7tretrieval source, not automatically a body citation · openreview.net
- https://arxiv.org/abs/2606.28356retrieval source, not automatically a body citation · arxiv.org
- https://arxiv.org/abs/2606.17443retrieval source, not automatically a body citation · arxiv.org
- https://searchengineland.com/fake-brand-ai-search-experiment-475947retrieval source, not automatically a body citation · searchengineland.com
- https://www.ktau.ai/research/training-time-brand-imprint-studyretrieval source, not automatically a body citation · www.ktau.ai
- https://arxiv.org/abs/2610.15033retrieval source, not automatically a body citation · arxiv.org
- https://www.gradial.com/blog/gradial-improved-ai-mentions-from-2-to-26-percentretrieval source, not automatically a body citation · www.gradial.com
- https://smokinggun.agency/our-work/geo-casestudy/retrieval source, not automatically a body citation · smokinggun.agency
- https://growthino.com/case-studies/from-zero-to-cited-by-ai-in-6-weeksretrieval source, not automatically a body citation · growthino.com
Evidence and qualifications
This is the existing redacted public record. It has not been rewritten for this page. Earlier privacy edits and collection qualifications continue to apply. Editorial text is English; a few original captures retain non-English UI badges or source-panel date labels. An answer’s empirical claims are platform output, not verified findings of this archive.
Body review
{
"reviewed_at": "2026-11-03T09:14:34.624328+00:00",
"reviewer": "assistant",
"complete_body_read": true,
"coherent_ending": true,
"annotation_and_explicit_url_coverage_checked": true,
"external_claims_verified": false,
"independent_human_double_coding": false
}
Existing public-text privacy policy
{
"body_changed": true,
"policy": "Only actual URL tracking, signed/session URL parameters and necessary identifying information are removed; original wording, including analytics parameter examples outside URLs, remains.",
"annotation_positions": "Original position indexes omitted because URL/privacy edits can change character locations."
}
Prompt SHA-256: 84f18194f3ea8dd08151f46577cb5614af08d775b76bd516de9d149a7fc01118
Answer SHA-256: 3b3e28201aad45515653cea2e878fed48a5394d31e833de0b8b375a1ed997cda