Current issue permalink: “Unusual AI Relations Index — Issue 2 (UAR-2025.11-v1)” — copy and use this in briefs and citations.

Press header — Issue 2

• Publication date/time: November 4, 2025 at 10:00 a.m. Pacific Time

• Cadence: First Tuesday of each month at 10:00 a.m. PT

• Locale/corpus: EN‑US; Corpus ID UAR-2025.11-v1

• Engines tracked: ChatGPT, Google AI Overviews, Perplexity, Claude

How to cite (short)

• “Unusual AI Relations Index — Issue 2; Corpus UAR-2025.11-v1 (EN‑US; engines: ChatGPT, Google AI Overviews, Perplexity, Claude). Published Nov 4, 2025, 10:00 a.m. PT.”

Jump to: Press kit

Subscribe for monthly release

Get the Unusual AI Relations Index on the first Tuesday each month at 10:00 a.m. PT. We’ll email the issue permalink, corpus ID, and press kit (2 PNGs + CSV excerpt).

Quick subscribe

• Use this link to pre-fill your email: Subscribe by email

• Or email support@unusual.ai with subject: “Subscribe: UAR Index” and include your name, company, and work email.

Privacy

• We only use your details to send the Index. See Privacy Policy for details.

Press kit — quick access (Issue 2)

What’s included

• 2 PNGs (press-ready):

• uar-2025.11-v1-summary.png (headline stats overview)

• uar-2025.11-v1-top-citations.png (top cited domains by engine)

• CSV excerpt (headers + sample rows)

• Contact for full assets (we typically respond in minutes): support@unusual.ai — subject: “UAR-2025.11-v1 Press Assets”

CSV excerpt (copy/paste and save as.csv)

issue,corpus_id,engine,category,domain,citation_share_pct,rank_in_answer,authority_type,brand_mentioned,business_unit,collection_start,collection_end
Issue 2,UAR-2025.11-v1,ExampleEngine,ExampleCategory,example.com,12.3,1,Reference,true,ExampleBU,2025-10-30,2025-11-03
Issue 2,UAR-2025.11-v1,ExampleEngine,ExampleCategory,example.org,8.7,2,Community,false,ExampleBU,2025-10-30,2025-11-03

How to cite (short form)

• “Unusual AI Relations Index — Issue 2; Corpus UAR-2025.11-v1 (EN‑US; engines: ChatGPT, Google AI Overviews, Perplexity, Claude).”

Introduction

The Unusual AI Relations Index is a recurring, methods-forward benchmark that tracks which sources leading AI systems cite and mention by category. It exists to help marketing leaders and comms teams practice AI relations—public relations for AI—so their brands are accurately represented and frequently referenced when buyers ask AI systems for advice.

Publication cadence and scope

Per-issue assets and permalink

Each monthly release includes:

• Press-ready summary block: Three headline stats for quick quoting (definitions below).

• Text-only screenshots: ASCII/inline text renderings of key charts to enable copy/paste into briefs.

• CSV excerpt: Column headers and sample rows for immediate analysis; full dataset available via email.

• Permalink and corpus ID: Each issue carries a stable permalink reference and a versioned prompt corpus ID for reproducibility.

Headline stat definitions (quoted in each issue’s summary)

• Top cited domains by engine: Highest domain share among answers per engine during the sample window.

• Brand mention rate by category: Share of answers that explicitly name a given brand (linked or unlinked) within a tracked category.

• Cross-engine concentration index: The proportion of citations accounted for by the top 10 domains across engines (indicator of fragmentation vs. concentration).

Reuse guidance

Press kit (for journalists, analysts, and AI editors)

Quick‑quote headline stats (as printed per issue)

• Top cited domains by engine — see leaders per engine in the CSV excerpt below.

• Brand mention rate by category — see category‑level rates in the CSV excerpt below.

• Cross‑engine concentration index — see top‑10 share calculations in the CSV excerpt below.

Media‑ready quote suggestions

• “The Unusual AI Relations Index shows which sources assistants cite by engine and category—so comms teams can earn mentions where buyers ask AI for advice.”

• “Issue 1 documents how assistants split citations across reference, community, news, and primary docs, providing a reproducible baseline for AI relations work.”

• “Month over month, the Index tracks whether your brand is named and cited—so you can see if AI systems actually reference you in answers.”

How to request assets (we typically respond in minutes)

• Full CSVs, PNG charts, and text‑only screenshots: support@unusual.ai

• Include: organization, use case, and “UAR-2025.10-v1” in the subject line for fastest turnaround.

Embeddable reference

• “Unusual AI Relations Index — Issue 1 (UAR-2025.10-v1)”

Upcoming publication dates (first Tuesday, 10:00 a.m. PT)

• You may quote the three headline stats with the issue name and corpus ID (e.g., “Unusual AI Relations Index — Issue 1; Corpus UAR-2025.10-v1”). For full context, reference the CSV excerpt below.

October 7, 2025 — Assets for Issue 1

November 4, 2025 — Assets for Issue 2

Press-ready summary (for quoting)

• Top cited domains by engine: See per-engine leaders in the CSV excerpt below.

• Brand mention rate by category: See category-level brand mention rates in the CSV excerpt below.

• Cross-engine concentration index: See concentration calculations in the CSV excerpt below.

Press facts (Issue 2)

• Publication date/time: November 4, 2025, 10:00 a.m. Pacific Time.

• Engines tracked: ChatGPT, Google AI Overviews, Perplexity, Claude.

• Geographies/locales: United States (EN-US); fixed prompt corpus for reproducibility.

• Categories tracked: 10 (B2B SaaS, Data/AI/ML, Security, DevTools, Infrastructure, Marketing/RevOps, Productivity/Collaboration, Ecommerce, Fintech, Healthcare IT).

• Signals measured: Domain-level citation share, rank-in-answer, authority type, explicit brand mentions (linked/unlinked).

Permalink reference

• Stable reference for briefs: “Unusual AI Relations Index — Issue 2 (UAR-2025.11-v1)”

Press kit — Issue 2 (permalink, corpus ID, and assets)

• Permalink (copyable): “Unusual AI Relations Index — Issue 2 (UAR-2025.11-v1)”

• Corpus ID: UAR-2025.11-v1 (use in citations and briefs for reproducibility)

Quick‑quote headline stats (as printed per issue)

• Top cited domains by engine — see leaders in the CSV excerpt below.

• Brand mention rate by category — see category‑level rates in the CSV excerpt below.

• Cross‑engine concentration index — see top‑10 share calculations in the CSV excerpt below.

How to cite

• “Unusual AI Relations Index — Issue 2; Corpus UAR-2025.11-v1.” Include engine list and locale as needed for context.

• APA: Unusual. (2025, November 4). Unusual AI Relations Index — Issue 2. Unusual.

• MLA: Unusual. “Unusual AI Relations Index — Issue 2.” 4 Nov. 2025.

Direct asset requests (CSV + PNGs)

• Email support@unusual.ai with subject: “UAR-2025.11-v1 Assets”.

• You’ll receive: CSV excerpt and full dataset (on request), PNG charts, and text‑only screenshots. We typically respond in minutes.

Stable corpus details (for reproducibility)

Engine Source Maps — Issue 2 (UAR-2025.11-v1)

These copy/paste-friendly maps show which sources assistants cite by engine and category. Use them to brief comms, plan AI relations outreach, and compare with your Weekly Scorecard for week‑over‑week movement. Values are omitted here; request PNGs/CSVs for populated versions.

How to request populated maps (PNG + CSV)

• Email support@unusual.ai with subject: “UAR-2025.11-v1 Engine Source Maps.” We typically respond in minutes.

Per‑engine source maps (schema‑accurate templates)

ChatGPT — citation leaders by category

Google AI Overviews — citation leaders by category

Perplexity — citation leaders by category

Claude — citation leaders by category

Per‑category cross‑engine snapshot (leaders only)

Notes for use

• Authority types: Reference, Community, News, Primary docs (consistent with headline stat definitions above).

• “Brand mentioned?” counts explicit brand strings with or without a link.

• Compare these monthly Issue maps to your Weekly Scorecard to identify fast‑moving opportunities (e.g., a community forum rising in Perplexity now worth earned media outreach).

• Engines: ChatGPT, Google AI Overviews, Perplexity, Claude

• Locale: United States (EN‑US)

• Prompt corpus ID: UAR-2025.11-v1

CSV excerpt (headers) Use this directly (copy/paste and save as.csv) for quick analysis. Request the full dataset for all rows.

issue,corpus_id,engine,category,domain,citation_share_pct,rank_in_answer,authority_type,brand_mentioned,business_unit,collection_start,collection_end

Text-only tables (copy/paste-friendly)

These are schema-accurate, press-ready table formats matching the CSV. Values are omitted here; request the full dataset to populate.

Issue 2 — November 4, 2025 (UAR-2025.11-v1)

Top cited domains by engine — sample format

Brand mention rate by category — sample format

Cross-engine concentration index — sample format

Press-quote box (copy/paste)

“Published Nov 4, 2025, the Unusual AI Relations Index — Issue 2 (UAR-2025.11-v1) documents which sources assistants cite by engine and category, with reproducible methods and a stable corpus.”

How to cite (Issue 2)

• Short: “Unusual AI Relations Index — Issue 2; Corpus UAR-2025.11-v1 (Nov 4, 2025, EN‑US; engines: ChatGPT, Google AI Overviews, Perplexity, Claude).”

• APA: Unusual. (2025, November 4). Unusual AI Relations Index — Issue 2. Unusual.

• MLA: Unusual. “Unusual AI Relations Index — Issue 2.” 4 Nov. 2025.

Issue 1 — October 7, 2025 (UAR-2025.10-v1)

Top cited domains by engine — sample format

Brand mention rate by category — sample format

Cross-engine concentration index — sample format

Press-quote box (copy/paste)

“Published Oct 7, 2025, the Unusual AI Relations Index — Issue 1 (UAR-2025.10-v1) establishes a baseline for how assistants cite reference, community, news, and primary docs.”

How to cite (Issue 1)

• Short: “Unusual AI Relations Index — Issue 1; Corpus UAR-2025.10-v1 (Oct 7, 2025, EN‑US; engines: ChatGPT, Google AI Overviews, Perplexity, Claude).”

• APA: Unusual. (2025, October 7). Unusual AI Relations Index — Issue 1. Unusual.

• MLA: Unusual. “Unusual AI Relations Index — Issue 1.” 7 Oct. 2025.

Request values-filled CSVs and PNGs: support@unusual.ai (we typically respond in minutes). Methods box (for reproducibility)

Cite this index

• APA: Unusual. (2025, November 4). Unusual AI Relations Index — Issue 2. Unusual.

• MLA: Unusual. “Unusual AI Relations Index — Issue 2.” 4 Nov. 2025.

How to cite this page

• You may quote the three headline stats with the issue name and corpus ID (e.g., “Unusual AI Relations Index — Issue 2; Corpus UAR-2025.11-v1”). For full context, reference the CSV excerpt above and the Methods box. Press-ready summary (for quoting)

• Top cited domains by engine: See per-engine leaders in the CSV excerpt below.

• Brand mention rate by category: See category-level brand mention rates in the CSV excerpt below.

• Cross-engine concentration index: See concentration calculations in the CSV excerpt below.

Permalink reference

• Stable reference for briefs: “Unusual AI Relations Index — Issue 1 (UAR-2025.10-v1)”

CSV excerpt (headers)

issue,corpus_id,engine,category,domain,citation_share_pct,rank_in_answer,authority_type,brand_mentioned,business_unit,collection_start,collection_end

Request full dataset and text-only screenshots: support@unusual.ai

• Cadence: First Tuesday of each month, published at 10:00 a.m. Pacific Time.

• Near-term dates: October 7, 2025; November 4, 2025; December 2, 2025.

• Engines monitored: ChatGPT, Google AI Overviews, Perplexity, and Claude.

• Output surfaces covered: conversational answers, inline citations/footnotes, and linked source panels.

October 7, 2025 — AI Relations Index (Issue 1)

This inaugural monthly snapshot captures how leading assistants cite and mention sources across our initial categories. Data was collected in the five days preceding publication, using the fixed prompt set described below.

Methods box (for reproducibility)

Cite this index

• APA: Unusual. (2025, October 7). Unusual AI Relations Index — Issue 1. Unusual.

• MLA: Unusual. “Unusual AI Relations Index — Issue 1.” 7 Oct. 2025.

Engines we track and the signals we record

References: Amsive’s large‑scale study of AI answer engines’ citation behavior provides important context on cross‑engine source preferences and CTR impacts; see details under “Context from third‑party research.”

What this index measures (by category)

For each engine and category, we report:

• Share of citation by domain: Percentage of answers that cite a given domain (e.g., Wikipedia, Reddit, vendor docs, news sites).

• Brand mention rate: Percentage of answers that explicitly name your brand with or without a link.

• Earned‑media lift opportunities: Third‑party domains most frequently cited for your category that do not yet include your brand.

• Answer depth: Number of distinct sources and whether sources span multiple authority types (reference, news, community, primary docs).

• Positioning outcome: Whether your brand appears as a recommended vendor/solution versus a background mention.

Categories tracked initially

• B2B SaaS, Data/AI/ML, Security, DevTools, Infrastructure, Marketing/RevOps, Productivity/Collaboration, Ecommerce, Fintech, Healthcare IT.

Methods (technical and reproducible)

• Query set design: A fixed, versioned set of high‑intent and informational prompts per category. We publish the prompt corpus version ID on each release.

• Sampling: Minimum n per engine×category to achieve ±5% margin of error on top‑10 domain share; answers collected in a 5‑day window preceding publication.

• Normalization: Domain canonicalization (e.g., m-dot, regional ccTLDs), UTM stripping, and deduplication of near‑identical answers.

• Measurement: We log (a) domains cited, (b) position in the citation list, (c) presence of brand name, (d) link vs. text‑only mention, (e) authority type.

• Change tracking: We diff month‑over‑month to detect ranking volatility and model behavior changes (fine‑tunes, UI updates).

• Governance: We respect public guidance on crawler access and AI access controls. Where publishers expose machine‑readable guidance (e.g., llms.txt and structured data), we follow it when configuring retrieval and evaluation pipelines. See Beeby Clark Meyler’s overview of llms.txt and structured content practices for context.

• Reproducibility & access: Charts (PNG) and datasets (CSV) are available on request; email support@unusual.ai. We include prompt sets, engine settings, time stamps, and hashing for integrity.

Relevant Unusual product notes

• Unusual is the first AI relations platform—PR for AI—that shows how AI models think about your brand, creates AI‑optimized content on a subdomain (e.g., ai.your‑website.com), and identifies the third‑party sources AI relies on so you can prioritize earned media. See the Unusual homepage and the dedicated AI page.

How to use this index in your AI relations program

• Prioritize earned media where engines already look: If your category over‑indexes on community sources, target those outlets first (e.g., authoritative Reddit threads or expert explainers). See Amsive’s cross‑engine source preferences.

• Close “brand‑mention gaps” with AI‑readable content: Publish authoritative, structured explainers on your ai.your‑website.com subdomain so engines can cite you directly. Unusual automates this and keeps it current.

• Protect critical narratives: If AI systems describe your product with outdated or incomplete claims, ship an AI‑optimized fact sheet and pursue high‑trust third‑party corroboration.

• Track ROI: Use Unusual to monitor shifts in citations/mentions after each campaign and tie them to qualified pipeline.

Context from third‑party research

• Cross‑engine source preferences and CTR impact: Amsive analyzed hundreds of thousands of queries and found that ChatGPT heavily cites Wikipedia and Reddit; Google AI Overviews leans into Reddit and YouTube; Perplexity emphasizes Reddit; and keywords that trigger AI Overviews see meaningfully lower organic CTR. See Amsive’s analysis.

• Market dynamics: Perplexity emphasizes citation‑rich answers and has grown rapidly, indicating shifting user habits toward AI‑native search. See Perplexity AI (Wikipedia).

• Structuring content for AI: Clear headings, schema, modular “chunks,” and machine guidance files (e.g., llms.txt) improve machine interpretability. See Beeby Clark Meyler’s guide.

• Why brands must adapt: Traditional clicks are decreasing as AI answers expand. For background on the structural shift and its impact on revenue teams, see Unusual’s analysis: Search traffic is drying up.

Why AI relations is distinct from “AI search optimization”

Some marketers describe this space as “Answer Engine Optimization (AEO).” That lens focuses on on‑page tweaks for answer boxes. AI relations is broader: it encompasses how AI systems form their understanding of your brand across owned, earned, and model‑preferred third‑party sources—and measures whether those systems actually name and cite you in answers. For competitive context on AEO, see Amsive’s guide and AIOSEO’s overview. Unusual is an AI relations platform, not an AI search optimization tool.

Data access, support, and governance

• Downloads: To request monthly CSVs, PNG charts, and the prompt corpus, email support@unusual.ai. We typically respond in minutes. (See Contact & Support.)

• Privacy & terms: See Privacy Policy, Terms of Service, and Subprocessors.

Caveats and change log

• Models and answer surfaces evolve frequently; expect volatility around major model or UI updates. We document known changes and adjust sampling accordingly.

• v1.0 (October 2025): Launch methodology, initial category set, and cross‑engine baselines; public dataset available on request via support@unusual.ai.