AI search citations: what do cited sources prove?

An AI search citation shows that a URL was attached to a specific API response or search action. It does not prove endorsement, factual correctness, stable visibility or ranking performance.

Read this after a native run when some sources appear beside a query and others remain at run level.

Inspect queries and cited sourcesRun one bounded provider search and keep query, action and run-level sources separate.Inspect a native result

First establish what the source link belongs to

A citation in a displayed answer connects a claim with a source. A search-action field may instead contain several sources for several queries without marking a sentence in the answer. An aggregated domain list is coarser still: it gives a host name, neither the exact URL nor the cited passage. Keep these three evidence types distinct in your notes.

Check whether the original response supplies a query-, search-action- or run-level relationship. A shared source field with four queries does not permit a unique join added afterwards. When an answer annotates a URL, preserve the permitted annotation and its scope; do not invent missing relationships because they seem plausible.

For the next step, use the OpenAI field guide.

A source link also carries no traditional rank position. Check the document to establish whether it supports the claim. Freshness, exact product version and the scope of a sentence matter more than the domain name alone. A recurring domain is not yet a recurring citation of the same page.

Check four original claims against the published example

Our worksheet annotates four owner-written sentences against best-seo-tools-us. It does not reconstruct an AI answer. The published normalized file contains query strings, domains and run metadata, not a full answer with resolvable citation passages. This exercise checks those data fields, not every sentence of a saved ChatGPT answer.

The first claim—four exposed queries in one search action—is supported by queries and usage.searchActions. The second—ahrefs.com appears in the domain set—can be checked in sourceDomains. It establishes no particular product claim. Nor can an exact Ahrefs URL be reconstructed from the domain name.

The third claim preserves providerResponseStatus = incomplete. The fourth asserts that a shared source supports every query; the file supplies no such relationship. The worksheet marks it unsupported. Rejecting that inference prevents a source occurrence from becoming an invented explanation of the answer.

Download the annotation and inspect the named fields yourself. Matthias Ramahi owns ai-fanout.com, SEO Fanout and Crawl Foundry. This published owner-run observation is a traceable first-party record, not independent product confirmation.

For the next step, use the URL and source audit.

Carry a supported claim into the right content

A price claim needs the current official pricing page with plan, currency, billing basis and review date. A search action containing the vendor domain does not supply all those conditions. Carry over only the supported part of the claim and retain open seat or licensing questions.

For the next step, use the worked product brief.

When evidence answers a specific reader question, add it to the page responsible for that task. Place a descriptive link beside the claim. A vendor document is not a function test you performed. Update or remove the statement when its conditions change.

Gemini also has rules for Grounded Results and Search Suggestions. This tool's local save filter removes those data. An empty source list in a later export therefore does not prove the original run had no sources. Record the retention scope before writing a comparison or report.

For the next step, use the Gemini storage explanation.

Common questions

These short answers cover recurring questions from tool use and source review. They are not claimed search-volume data.

What is a ChatGPT citation?: It is a source link attached to a displayed ChatGPT answer. An OpenAI API citation is evidence from a different surface: it belongs to the named API response and must not be presented as a permanent ChatGPT ranking.
What is a Gemini citation?: It is source or grounding information returned at the scope supplied by a Gemini response. Google's product terms and response structure determine how that data can be shown, stored and reused.
Does an AI citation mean the page ranks number one?: No. A citation records that a URL appeared in the supported scope of one provider response. It contains no traditional organic rank, search volume or guarantee of recurrence.
Can every cited source be assigned to one query?: Only when the provider response supplies that relationship. Sources from a multi-query action or a run-level citation list must stay at that broader scope.

Sources used

These sources support the functions and limits described here. They do not prove claims beyond that scope.

Original annotated workflow, controls, prompts and export exercises
Original diagrams and editorial annotations; only the previously approved normalized OpenAI records. Synthetic selection files use actual producers and independent CSV readback, not new provider output or Excel/Sheets import. Own dated public-page checklist makes indexing and outcomes explicitly unproven.
Open source
OpenAI web search tool guide
Documents web_search_call output and search actions that usually, but not always, include the query or queries searched. Search actions, exposed strings and cited sources therefore need separate interpretation. It does not expose chain of thought or guarantee that every route returns query strings. Rechecked for the original 2026-10-10 zero-query decision table; no new provider observation is published.
Open source
Gemini grounding with Google Search
Documents Gemini 3.7 Flash Google Search, google_search_call arguments.queries, cited URL annotations and per-query billing.
Open source
Gemini API Additional Terms
Restricts caching, syndication, analysis and reuse of Google Grounded Results and Search Suggestions; drives the Gemini local-save redaction and public-example boundary.
Open source
Dated OpenAI fanout example observations
Four owner-run OpenAI API observations record exact inputs, timestamps, exposed query strings, search-action source domains, usage, method versions and response status. Versioned JSON, normalized CSV and a JSON Schema are published. They are not an independent benchmark.
Open source