> ## Content Index > Fetch the complete content index at: https://www.promptthemarket.com/llms.txt > Use this file to discover other available public pages before exploring further. # OpenAI's own API misses what ChatGPT cites - URL: https://www.promptthemarket.com/openai-api-misses-what-chatgpt-cites/ - Published: 2026-10-19T09:00:00.000Z - Updated: 2026-10-19T09:00:00.000Z - Description: Researchers say Google and the company’s interface together identified 26-28% of referenced pages in an 11 October 2026 preprint, not peer reviewed. - Author: Editor News - Tags: Research, Australia, OpenAI, #author-james-dore ![James Dore](https://storage.ghost.io/c/e4/01/e4017178-c1da-4131-80aa-35f02cb5b999/content/images/2026/09/james-dore.jpg) By **James Dore** · 2026-10-19 TLDR Google Search and OpenAI's Responses API together matched fewer than 28 per cent of exact pages ChatGPT cited in real signed-in sessions, an arXiv preprint published 11 October 2026 found. Matching by website reached around 50 per cent. Results also varied run to run: 44.5 per cent of distinct cited pages appeared in only one of 20 repeated runs. KEY TAKEAWAYS 01Google Search alone matched just 6.4 per cent of exact URLs cited by GPT-5.5, the study found. 02Combined proxies recovered 26 to 28 per cent of cited pages across both model versions tested. 0344.5 per cent of distinct cited pages appeared in only one of 20 repeated GPT-5.5 runs. 04ChatGPT never searched with the question's exact wording, and mostly searched in English for Chinese questions. 05The unreviewed arXiv preprint recommends separating retrieval visibility from citation visibility. ## The gap between the stand-ins and ChatGPT Tools that estimate how often ChatGPT cites a brand cannot see inside a user's ChatGPT session, so they lean on stand-ins: Google results, or OpenAI's Responses API with web search switched on. A preprint from researchers at the National University of Singapore and Xi'an Jiaotong University checked both against what ChatGPT actually cited in real signed-in sessions.[\[2\]](https://www.searchenginejournal.com/openais-api-only-partly-matched-chatgpts-citations-paper-finds/?ref=promptthemarket.com) Combined, Google and the Responses API matched 28.3 per cent of the pages GPT-5.3 Instant cited and 26.2 per cent for GPT-5.5 Instant. Google on its own matched 9.4 and 6.4 per cent.[\[2\]](https://www.searchenginejournal.com/openais-api-only-partly-matched-chatgpts-citations-paper-finds/?ref=promptthemarket.com) Roughly three in four cited pages turned up in neither stand-in. Matching by website rather than by page did better: 50.3 per cent for GPT-5.3 and 48.7 per cent for GPT-5.5.[\[2\]](https://www.searchenginejournal.com/openais-api-only-partly-matched-chatgpts-citations-paper-finds/?ref=promptthemarket.com) That is the difference between knowing a brand's site was in the mix and knowing which article ChatGPT actually pointed people to. Share of pages ChatGPT (GPT-5.5 Instant) cited that each stand-in also found Google results alone6.4% OpenAI Responses API sources22.5% Google and API combined26.2% Combined, matched by site not page48.7% Averages across 80 questions. Source: Jiang et al., arXiv 2610.15382, via Search Engine Journal. ## How the study was run Between 29 March and 29 July 2026 the researchers recorded the data flowing through their own signed-in ChatGPT sessions to see what it searched for, what came back and what it cited. They used 80 questions from Search Arena, 40 in English and 40 in Chinese, and ran GPT-5.3 Instant and GPT-5.5 Instant 20 times on each one.[\[2\]](https://www.searchenginejournal.com/openais-api-only-partly-matched-chatgpts-citations-paper-finds/?ref=promptthemarket.com) The answers moved around a lot. For GPT-5.5, 44.5 per cent of the distinct pages cited for a question showed up in only one of the 20 runs.[\[2\]](https://www.searchenginejournal.com/openais-api-only-partly-matched-chatgpts-citations-paper-finds/?ref=promptthemarket.com) In runs where its searches were visible, GPT-5.3 and GPT-5.5 never searched using the question's exact wording, and on Chinese questions around two-thirds of runs searched only in English.[\[2\]](https://www.searchenginejournal.com/openais-api-only-partly-matched-chatgpts-citations-paper-finds/?ref=promptthemarket.com) ## What it means for AI visibility reporting For Australian brands paying for AI share-of-voice reports, the study is a caution about method rather than a verdict on any product: it tested no commercial tracking tool.[\[2\]](https://www.searchenginejournal.com/openais-api-only-partly-matched-chatgpts-citations-paper-finds/?ref=promptthemarket.com) A report built on one run of each prompt, typed exactly as a customer would, against Google or the API, is measuring something that only partly overlaps with what ChatGPT shows. The authors argue generative engine optimisation should treat being retrieved and being cited as separate measures, allow for ChatGPT rewriting queries, and track sites and entities as well as exact URLs.[\[1\]](https://arxiv.org/abs/2610.15382?ref=promptthemarket.com) Search Engine Journal adds that every setting in the paper predates GPT-5.6 and GPT-6, so any method should be retested as models change.[\[2\]](https://www.searchenginejournal.com/openais-api-only-partly-matched-chatgpts-citations-paper-finds/?ref=promptthemarket.com) The paper is a preprint posted to arXiv on 11 October and has not been peer reviewed.[\[1\]](https://arxiv.org/abs/2610.15382?ref=promptthemarket.com) Related coverage - [GPT-6 puts charts and calculators inside ChatGPT's answers](https://www.promptthemarket.com/gpt-6-puts-charts-and-calculators-inside-chatgpt-answers/) - [ChatGPT's AI search share falls from 76% to 53% as Gemini gains](https://www.promptthemarket.com/chatgpt-s-ai-search-share-drops-from-76-to-53-as-gemini-surges/) - [GPT-6 makes ChatGPT visits look direct in site analytics](https://www.promptthemarket.com/gpt-6-stripped-chatgpt-s-tracking-tags-hiding-31-000-visits-a-day/) - [Google's AI answers cost publishers 40% of their clicks, trial finds](https://www.promptthemarket.com/ai-overviews-cut-publisher-clicks-by-40-pay-rate-secret/) SOURCES & CITATIONS 1. [arXiv: Understanding ChatGPT Citations: Evidence from Production Web Search Across Model Generations](https://arxiv.org/abs/2610.15382?ref=promptthemarket.com) 2. [Search Engine Journal: OpenAI's API only partly matched ChatGPT's citations, paper finds](https://www.searchenginejournal.com/openais-api-only-partly-matched-chatgpts-citations-paper-finds/?ref=promptthemarket.com) ![James Dore](https://storage.ghost.io/c/e4/01/e4017178-c1da-4131-80aa-35f02cb5b999/content/images/2026/09/james-dore.jpg) [James Dore](https://www.promptthemarket.com/james-dore/) James Dore is founder and editor of Prompt the Market and Strategy Director at Bushnote, a Sydney strategy, SEO and AI search agency. He has a background in strategy, media and financial services, and led consumer product marketing at MONEYME. He writes on AI search, answer engines, technology policy and media economics.