# Will AI Assistants Cite Articles Drafted by a Ghostwriter or by AI?

> AI assistants cite a page for what it says and whether they can read it, not for who drafted it. What to add and check before a ghostwritten article goes out.

Published: 2026-11-20
Source: https://www.itgrows.ai/blog/do-ai-assistants-cite-ghostwritten-articles

---

<p>Yes, they can: an assistant cites a page for what it says and whether it can fetch and read it, and Google states that it rewards quality content however it is produced. Whether they do depends on what the article adds that only you could have written, your examples, your numbers and your opinions, and on whether you checked every line before it went out under your name.</p>
<h2>Does an assistant know or care who typed the article?</h2>
<p>It sees a page, not the process behind it. OpenAI's crawler documentation describes two agents that may touch your article: <a href="https://developers.openai.com/api/docs/bots" target="_blank" rel="noopener">OAI-SearchBot, which is used to surface websites in ChatGPT's search results, and ChatGPT-User, which may visit a web page when a user asks ChatGPT a question</a>. Perplexity describes the same pattern: when users ask a question, <a href="https://docs.perplexity.ai/docs/resources/perplexity-crawlers" target="_blank" rel="noopener">Perplexity-User might visit a web page to help provide an accurate answer and include a link to the page in its response</a>. Both read the text that is on the page when they arrive. A byline is part of that text; who sat at the keyboard is not.</p>
<p>We see these visits in server logs. Our own course marketplace (pickaclass.com) received 1,918 fetches on content pages by ChatGPT-User and Perplexity-User (sent while an assistant answers a person) and OAI-SearchBot (ChatGPT's search crawler), 28 July – 11 August 2026 (its server log); a client blog of a few dozen articles received 56 fetches by the same agents across 24 different articles. Only the ChatGPT-User and Perplexity-User fetches happen while an assistant is answering a person; OAI-SearchBot crawls for ChatGPT's search. Nothing in a log line records how an article was drafted.</p>
<p>Google set out the same position for its search results on 8 February 2023. Its guidance on AI-generated content opens with a section titled <a href="https://developers.google.com/search/blog/2023/02/google-search-and-ai-content" target="_blank" rel="noopener">"Rewarding high-quality content, however it is produced"</a>, says that appropriate use of AI or automation is not against its guidelines, and puts the point plainly: using AI doesn't give content any special gains, because "It's just content."</p>
<p>So "who drafted this?" is the wrong question. The useful ones are: can an assistant reach the page, and once it reads it, is there anything on it worth quoting?</p>
<h2>Why would an assistant pass over a ghostwritten article?</h2>
<p>Usually because it repeats what is already online, and an assistant that searches has other pages to quote. In a test we ran on 23 September 2026 with 20 German business questions, the model answered 10 without searching at all and cited nobody; where it searched, it cited primary sources (laws, ministries, OECD, large accounting firms) on 9 of 10; firms were named only on 'who should I hire / how to choose' questions; two runs of the same question named two different sets of firms. The pages that got cited were the ones where the information starts.</p>
<p>A draft produced from a topic alone, by a freelancer or by a model, is close to a rewrite of pages that already exist. Google's page on <a href="https://developers.google.com/search/docs/fundamentals/creating-helpful-content" target="_blank" rel="noopener">creating helpful, reliable, people-first content</a> asks whether content provides "original information, reporting, research, or analysis", and whether content that draws on other sources avoids "simply copying or rewriting those sources". The same page lists what Google's quality raters look at. Under effort it says that "using generative AI to produce large amounts of text without manual oversight or curation represents little to no effort"; under originality it asks for information or perspectives "that aren't already available on other websites". It adds that attribution or giving credit to other sources doesn't replace the need for original effort.</p>
<p>Neither a ghostwriter nor a drafting tool can supply that originality alone. They don't know which client nearly walked away and why, which pricing model you dropped, or where you think the standard advice in your field is wrong. You do.</p>
<h2>What must you add before it goes out under your name?</h2>
<p>The parts that make the article yours, and therefore worth citing:</p>
<ul>
<li><strong>An example from your own work.</strong> A real situation, anonymised where needed: what the client asked, what you did, what happened next. One specific case says more than three paragraphs of general advice.</li>
<li><strong>Numbers you can stand behind.</strong> Your own counts, prices, durations or results, with where they come from. If you have no number, describe the mechanism instead of letting the draft invent one.</li>
<li><strong>An opinion.</strong> Where you disagree with the usual advice in your field, and why. An assistant answering a "how should I…" question needs positions to report, and a generic draft rarely takes one.</li>
<li><strong>A direct answer at the top.</strong> The first two sentences should answer the question in the title, so a reader, or an assistant reading the page, gets the point without scrolling.</li>
<li><strong>A byline that leads somewhere.</strong> Google's people-first page asks whether pages carry a byline where one might be expected and whether bylines lead to further information about the author. An author page that lists your work and links your profiles does that job.</li>
</ul>
<p>If you have never looked at what the assistants already say about you, start there: <a href="https://www.itgrows.ai/blog/how-to-check-what-ai-assistants-say-about-you">How to Check What AI Assistants Say About You</a> walks through it.</p>
<h2>What must you check before you publish?</h2>
<p>Everything, and above all anything a model drafted. Google's <a href="https://developers.google.com/search/docs/fundamentals/using-gen-ai-content" target="_blank" rel="noopener">guidance on using generative AI content</a> (last updated 1 October 2026) explains why: generative models don't retrieve facts, they predict a likely sequence of words based on their training data, so their output may contain inaccuracies. Google calls it critical to manually fact-check and review all AI-generated content for accuracy and trustworthiness before publishing, and extends that review to titles, meta descriptions, structured data and image alt text.</p>
<p>A ghostwriter's draft needs the same pass for a different reason: they were not in the room for your work. Before an article goes out under your name, check that:</p>
<ol>
<li>Every number has a source you could show someone, and every claim about your own work is true as written.</li>
<li>No quote, client story or study was invented to fill a gap.</li>
<li>Every link opens and points to the page it claims to.</li>
<li>The title says what the page answers, without exaggeration.</li>
<li>The article sits on the page as plain text, not inside an image or a scanned PDF.</li>
<li>Assistants are allowed to reach it. OpenAI notes that sites opted out of OAI-SearchBot will not be shown in ChatGPT search answers, so a blanket block on crawlers removes the article from those answers however good it is.</li>
</ol>
<p>If an assistant repeats a wrong figure from your article, the correction has to start on your page, so it is cheaper to catch it before publishing. For posts on LinkedIn rather than your own site, what assistants can read there is a separate question, covered in <a href="https://www.itgrows.ai/blog/does-chatgpt-read-linkedin-profiles">Does ChatGPT Read LinkedIn Profiles and Posts?</a></p>
<h2>Should the byline say that a ghostwriter or AI helped?</h2>
<p>Google's guidance does not mention ghostwriters. What it asks for is authorship that is accurate and not deceptive. The <a href="https://developers.google.com/search/blog/2023/02/google-search-and-ai-content" target="_blank" rel="noopener">February 2023 post on AI-generated content</a> recommends accurate author bylines where readers would reasonably expect them, suggests AI or automation disclosures for content where someone might wonder "How was this created?", and says that giving AI an author byline is probably not the best way to make clear that AI was part of the process.</p>
<p>The people-first page draws the hard line. Fabricating creator profiles, such as AI-generated headshots, made-up names or false credentials that make content appear written by human experts, is deception, and it makes a page untrustworthy to users and to Google's automated systems. The page also asks you to think about the "Who, How, and Why" of your content, and it calls "Why" perhaps the most important: content made primarily to help people, not to attract search visits.</p>
<p>That leaves a workable rule. Put your name on an article when the experience, the examples and the opinions in it are yours and you approved the final text. If a reader would reasonably wonder how it was produced, say so in one line. Never put an invented expert on a page.</p>
<h2>Where do a ghostwriter or a drafting tool fit?</h2>
<p>At the stage they are good at: structure and first drafts. Google's generative AI guidance says such tools can be particularly useful when researching a topic and to add structure to original content, and warns that using them to generate many pages without adding value for users may violate its spam policy on scaled content abuse. The same split works with a human ghostwriter: they interview you, shape your material and draft; you supply the substance and sign off.</p>
<p>A setup that holds up:</p>
<ul>
<li><strong>Feed the drafter material, not just a topic.</strong> Voice notes, a client email with the names removed, the slide you showed at a workshop, your answer to a question a prospect asked on a call.</li>
<li><strong>Keep the volume to what you can review.</strong> One article you have checked line by line is worth more than ten you skimmed.</li>
<li><strong>Edit for your claims and your voice, not only for typos.</strong> If a sentence says something you would not say out loud to a client, cut it.</li>
</ul>
<p>Agencies that do this for several clients run the same loop at scale; <a href="https://www.itgrows.ai/blog/how-agencies-run-linkedin-ghostwriting-for-clients-mvi4fphd">How Agencies Run LinkedIn Ghostwriting for Clients</a> describes the workflow. For posts on LinkedIn and X, the <a href="https://www.itgrows.ai/personal">ItGrows LinkedIn product</a> drafts posts in your own voice on topics you choose; you approve or edit every post, and it publishes on your schedule through the platforms' official APIs. If you run a company, <a href="https://www.itgrows.ai/">ItGrows</a> makes businesses visible in AI answers and measures every week whether ChatGPT, Claude, Gemini and Perplexity name them.</p>
<h2>Frequently Asked Questions</h2>
<p><strong>Does Google penalise articles written with AI?</strong></p>
<p>Not for being written with AI. Google's guidance says appropriate use of AI or automation is not against its guidelines. Using automation, including AI, to generate content with the primary purpose of manipulating search rankings is what violates its spam policies.</p>
<p><strong>Can I list AI as the author of my article?</strong></p>
<p>Google says giving AI an author byline is probably not the best way to make clear that AI was part of the content creation process. If a reader would reasonably wonder how the article was made, a short disclosure line does that job better.</p>
<p><strong>Do I have to say that a ghostwriter helped?</strong></p>
<p>Google's guidance does not mention ghostwriters; it asks for accurate bylines and warns against deceptive authorship such as made-up names or false credentials. Publications that run guest articles may set their own rules, so check those before you submit.</p>
<p><strong>How do I find out whether assistants cite my articles?</strong></p>
<p>Ask ChatGPT, Claude, Gemini and Perplexity the questions your articles answer, in the words a client would use, and note which pages they link. Repeat the same questions on a schedule, because one answer is a snapshot; <a href="https://www.itgrows.ai/blog/how-to-get-chatgpt-to-recommend-you-as-an-expert">How to Get ChatGPT to Recommend You as an Expert</a> covers what to change when your pages are missing.</p>