Guide

Does republishing on Medium, LinkedIn or Substack hurt your AI citations?

Should your article also go on Medium, LinkedIn or Substack? We read Google's and Bing's guidance and the three platforms' help pages and robots.txt files as of November 11, 2026. Nobody penalizes you, but the platform copy can take the citation your own page should have earned.

Quick answer

Not through a penalty. Google says there is no duplicate content penalty, but it shows one copy of duplicated text and may pick the platform's. Google tells syndication partners to noindex copies, not rely on canonical tags. Medium adds a canonical on import; LinkedIn and Substack document none. And each platform admits different AI crawlers.

What are the key facts about republishing and AI citations?

FactDetailSource
No duplicate content penalty"There's no such thing as a 'duplicate content penalty'"; action only if the intent is "to be deceptive and manipulate search engine results" (2008)Google
Duplicates are not spam"Some duplicate content on a site is normal and it's not a violation of Google's spam policies"Google
Google shows one copyGoogle groups duplicates, picks a canonical, and "a Google Search result usually points to the canonical page"Google
Canonical is only a hintGoogle calls it "a hint, not a rule" and may pick a different pageGoogle
Syndicated copiesThe canonical link element "is not recommended" for syndication; "the most effective solution is for partners to block indexing." Added May 2, 2023Google
Google AI Overviews and AI ModeA page must be "indexed and eligible to be shown in Google Search with a snippet"; no extra requirementsGoogle
Bing and CopilotDuplicate URLs "reduce Bing's confidence in selecting a URL for grounding results or citations"Bing
MediumThe import tool adds "a canonical URL referencing the original source material" automaticallyMedium
LinkedIn articlesArticle settings offer an SEO title and SEO description; no canonical field is documentedLinkedIn
SubstackNo canonical setting in the help pages we read; a toggle tells AI tools not to train on your postsSubstack
AI crawler accessMedium blocks GPTBot and ClaudeBot; LinkedIn blocks GPTBot, ClaudeBot and PerplexityBot; Substack names no AI crawlerMedium, LinkedIn, Substack robots.txt
OpenAI, Anthropic, PerplexityTheir crawler pages say nothing about duplicate copies or canonical tagsOpenAI, Anthropic, Perplexity

Is there a duplicate content penalty for republishing your own articles?

No. Google wrote in 2008: "There's no such thing as a 'duplicate content penalty.'" It said duplicate content is only grounds for action when "the intent of the duplicate content is to be deceptive and manipulate search engine results." Instead, "Google tries to filter out duplicate documents so that users experience less redundancy." Google That post is 18 years old, but Google's current docs agree: "Some duplicate content on a site is normal and it's not a violation of Google's spam policies." Google

The real cost is the filter, not a punishment. When Google finds the same text at several URLs, it groups them and picks one, the canonical. "The canonical page will be crawled most regularly; duplicates are crawled less frequently," and "a Google Search result usually points to the canonical page." Google If your article sits on your site and on Medium, one usually gets shown. The other mostly disappears from results.

Bing's current guidelines agree. They which cover "Bing search experiences, Copilot, and grounding API results." Publishing the same content across URLs "reduces confidence in selecting a preferred version." Bing

In plain words: you won't be punished for posting your article on Medium. You may just stop being the copy that gets shown and cited.

Which copy gets cited in Google AI Overviews and AI Mode?

Whichever copy Google has indexed and can show with a snippet. Google's AI features page says a supporting link must be "indexed and eligible to be shown in Google Search with a snippet," and "there are no additional technical requirements." Google Google's AI page doesn't mention duplicates. But a filtered-out duplicate is not the URL Google normally shows, so in practice the canonical copy is the one in line to be cited.

You don't get the final say. Google picks the page that "is objectively the most complete and useful for search users," and "Google may choose a different page as canonical than you do." Google Its troubleshooting page adds that, rarely, the algorithm "may select a URL from an external site that is hosting your content." Google

Google's advice for syndicated content is specific. "The canonical link element is not recommended for those who want to avoid duplication by syndication partners, because the pages are often very different. The most effective solution is for partners to block indexing of your content." Google Google's changelog dates this: on May 2, 2023 it "updated our documentation about canonicalization with explicit recommendations for syndicated content." Google

There is a reporting catch. If Google picks a canonical on a site you don't own, "you won't be able to see any of the traffic for your duplicate page" in Search Console. Google So if the Medium copy wins, the Search Console data for that copy sits in Medium's property, not yours.

Can AI crawlers read your copy on Medium, LinkedIn and Substack?

Some can, and the mix differs by platform. We opened each platform's robots.txt on November 11, 2026. A crawler that isn't named in the file follows the general User-agent: * rules.

CrawlerWhat it is forMediumLinkedInSubstack
GPTBotOpenAI model trainingBlocked by nameBlocked by nameNot named; general rules allow posts
OAI-SearchBotChatGPT searchNot named; general rules allow storiesNamed; its own rules do not block article pagesNot named; allowed
ChatGPT-UserPages a ChatGPT user asks forNot named; allowedBlocked by nameNot named; allowed
ClaudeBotAnthropic model trainingBlocked by nameBlocked by nameNot named; allowed
Claude-SearchBotClaude searchNot named; allowedNamed; its own rules do not block article pagesNot named; allowed
Claude-UserPages a Claude user asks forNot named; allowedBlocked by nameNot named; allowed
PerplexityBotPerplexity searchNot named; allowedBlocked by nameNot named; allowed
Perplexity-UserPages a Perplexity user asks forNot named; allowedBlocked by nameNot named; allowed
Google-ExtendedGemini training and Gemini app groundingNot named; allowedBlocked by nameNot named; allowed

Sources: Medium, LinkedIn, Substack.

What the bots do comes from the vendors. OpenAI says OAI-SearchBot "is used to surface websites in search results in ChatGPT's search features." OpenAI Anthropic says blocking Claude-SearchBot "may reduce your site's visibility and accuracy in user search results." Anthropic Perplexity says PerplexityBot is "designed to surface and link websites in search results on Perplexity." Perplexity Google-Extended "does not impact a site's inclusion in Google Search," so LinkedIn blocking it does not keep LinkedIn articles out of AI Overviews. Google

Three things follow:

  • Medium blocks the training bots but not the search bots. ChatGPT search, Claude search and Perplexity can read a Medium copy, the same engines you want reading your site.
  • LinkedIn opens its article pages to OAI-SearchBot, Claude-SearchBot, Googlebot and Bingbot, and blocks Perplexity's bots. The file also opens with a warning that "The use of robots or other automated means to access LinkedIn without the express permission of LinkedIn is strictly prohibited." LinkedIn Our LinkedIn and AI answers guide covers how often engines cite LinkedIn.
  • Substack names no AI crawler at all. A Substack newsletter usually runs on its own subdomain or custom domain, so check that address's /robots.txt as well.

A robots.txt file is a request, not a lock, and doesn't show other blocking a platform may do. Treat "allowed" here as "not refused by the file." Our guide to social media posts and AI recommendations runs the same check for Instagram, Facebook, TikTok, X and YouTube.

Do Medium, LinkedIn and Substack let you point back to the original?

Only Medium documents a way. The other two don't, as far as their help pages show.

Medium. Its import tool "will automatically backdate the post to the original date, as well as add a canonical link." Medium adds that "Imported stories will automatically apply a canonical URL referencing the original source material," and that you may only import content you have the right to. Medium Remember that Google treats a canonical as a hint, and says it is not the recommended fix for syndicated copies. Google

LinkedIn. The publishing settings let you add an "SEO title" and an "SEO description" that control how an article or newsletter appears on search engines. LinkedIn We found no canonical setting there or elsewhere in LinkedIn's help. Treat a LinkedIn article as a separate page Google judges on its own.

Substack. We found no canonical setting in Substack's help center. It does document a privacy toggle, "Tell AI tools not to train their models on your content." Substack warns that this "will only apply to AI tools that respect this setting" and that "blocking training may limit your publication's discoverability in tools and search engines that return AI-generated results." Substack

None of the three platforms' help pages we read describes a way to noindex a single post, the fix Google recommends for syndicated copies. So on all three, your copy is likely to be indexed and compete with yours.

What do ChatGPT, Claude, Perplexity and Copilot say about duplicate copies?

OpenAI, Anthropic and Perplexity say nothing. Their crawler pages, read on November 11, 2026, don't mention canonical tags, duplicate content, syndication or republished copies. OpenAI Anthropic Perplexity How ChatGPT, Claude or Perplexity choose between your page and a Medium copy of it is unconfirmed. Anyone who tells you these engines "respect canonical tags" is guessing.

What OpenAI does document is the access rule: "Sites that are opted out of OAI-SearchBot will not be shown in ChatGPT search answers, though can still appear as navigational links." OpenAI So a copy on a platform that blocks OAI-SearchBot can't win a ChatGPT search citation, and a copy on a platform that allows it can.

Bing is the one AI search company with a clear written view, and it covers Copilot: "Duplicate URLs dilute signals and reduce Bing's confidence in selecting a URL for grounding results or citations." They also say "Canonical tags help signal preference but do not replace fixing underlying issues." Bing Bing's rule on scraped content targets copying "from other sites without adding original value." It doesn't address authors republishing their own work.

For how each engine finds its sources in the first place, see where AI engines get their sources.

How do I republish without losing the AI citation?

Our advice, not a vendor statement: publish at home first, prove Google picked your page, then share a clearly different version that links back.

  1. Publish the full article on your own site first. Give it a self-referencing canonical and a visible date. This is the copy you want cited.
  2. Wait until Google has indexed it. Use the URL Inspection tool in Search Console to check "which page Google considers canonical." Google Don't republish until it shows your own URL.
  3. On Medium, use the import tool, not copy and paste. Import adds the canonical link and backdates the story. Medium Better, import a shorter version, since a canonical is only a hint.
  4. On LinkedIn, write a different article on the same topic. With no canonical field, a word-for-word copy just competes with you. Google says pages split out of a duplicate cluster faster "if the difference between the new content and the other clustered pages is clear and significant." Google Answer a narrower question, and link to the full piece in the first paragraph.
  5. On Substack, send the summary and link to the full article. Substack documents no canonical setting, so a full copy is a second indexable page.
  6. Keep money pages unique. Never republish service, pricing or location pages. You most need AI to cite those on your own domain.
  7. Add an "originally published at" line with a link. It tells readers and engines where the original lives.
  8. Check again in two to four weeks. Google says it "might hold pages in a duplicate cluster for up to two weeks" even after you fix content issues. Google Then ask the engines your question and see which URL they cite.

What should be on my republishing checklist?

CheckDone whenSource
Original is live firstThe full article is on your own domain with a visible dateOur advice
Google picked your pageURL Inspection shows your own URL as the Google-selected canonicalGoogle
Bing has it tooYour page is indexed in Bing Webmaster Tools before you republishOur advice
Medium copy importedThe story came in through Import a story, so it carries a canonical to your pageMedium
LinkedIn copy is differentThe article answers a narrower question and links to your full pieceLinkedIn; Google
Substack sends a summaryThe newsletter carries the key points and a link, not the full textOur advice
Money pages not copiedService, pricing and location pages exist only on your siteOur advice
Crawler access checkedYou know which AI crawlers each platform's robots.txt lets inMedium, LinkedIn, Substack robots.txt
Back-link on every copyEach copy says where the original lives, with a linkOur advice
Citations re-checkedTwo to four weeks later, you know which URL each engine citesGoogle

How do I see which copy AI engines are citing?

Ask the engines and look at the sources they list. For Google, use URL Inspection to see which page it treats as canonical. If it names the Medium or LinkedIn copy, rework that copy so it is clearly different, then use "Request Indexing" on your own page. Google notes the feature "is subject to quotas," so save it for your important URLs. Google

ChatGPT, Claude, Perplexity and Copilot have no console showing which copy they prefer. The only test is to ask. Our Prompt Simulator runs one question across the engines and lists every source each one cites, so you can see whether the link goes to your site, Medium, LinkedIn or someone else. Our free AI visibility check shows where the engines already mention your business and where they don't.

If a platform copy keeps winning, don't delete your page. Make yours the most complete answer, which Google says it picks, and shorten the platform version. Google

Sources

  1. Fix canonicalization issues (Syndicated content, URL Inspection), Google Search Central. Read Nov 11, 2026.
  2. Latest Google Search documentation updates (May 2, 2023 entry), Google Search Central. Read Nov 11, 2026.
  3. What is URL canonicalization, Google Search Central. Read Nov 11, 2026.
  4. Demystifying the "duplicate content penalty" (September 2008), Google Search Central Blog. Read Nov 11, 2026.
  5. AI features and your website, Google Search Central. Read Nov 11, 2026.
  6. Google's common crawlers (Google-Extended), Google Search Central. Read Nov 11, 2026.
  7. Bing Webmaster Guidelines, Microsoft Bing. Read Nov 11, 2026.
  8. Overview of OpenAI Crawlers, OpenAI. Read Nov 11, 2026.
  9. Does Anthropic crawl data from the web, and how can site owners block the crawler?, Anthropic. Read Nov 11, 2026.
  10. Perplexity Crawlers, Perplexity. Read Nov 11, 2026.
  11. Importing a post to Medium, Medium Help Center. Read Nov 11, 2026.
  12. medium.com robots.txt, Medium. Read Nov 11, 2026.
  13. linkedin.com robots.txt, LinkedIn. Read Nov 11, 2026.
  14. SEO settings for articles or Newsletters published on LinkedIn, LinkedIn Help. Read Nov 11, 2026.
  15. substack.com robots.txt, Substack. Read Nov 11, 2026.
  16. How can I block AI from using my Substack publication to train their models?, Substack Support. Read Nov 11, 2026.
FAQ

Common questions

Will Google penalize me for posting my blog article on Medium or LinkedIn?

No. Google says there is no duplicate content penalty unless the intent is to deceive, and that duplicate content is not a violation of its spam policies. It does filter duplicates, so only one copy usually shows. Google Google

Does Medium add a canonical link to imported stories?

Yes. Medium's import tool backdates the story and automatically applies a canonical URL pointing to the original. Google treats canonicals as a hint, not a rule. Medium Google

Can I set a canonical URL on a LinkedIn article?

Not according to LinkedIn's help. Article settings offer an SEO title and SEO description only. Publish a reworked version that links to your original instead. LinkedIn

Can ChatGPT and Perplexity read my Medium or LinkedIn copy?

On Medium, yes: its robots.txt blocks GPTBot and ClaudeBot but not OAI-SearchBot or PerplexityBot. On LinkedIn, OAI-SearchBot may read article pages but PerplexityBot is blocked by name. Medium LinkedIn

Should the republished copy use a canonical tag or noindex?

Google says canonical tags are not recommended for syndicated copies and that blocking indexing works best. None of the help pages we read from Medium, LinkedIn or Substack offers a per-post noindex, so publish a shorter or different version instead. Google

See which copy of your article AI cites.

Enter your domain. The free AI visibility check shows where ChatGPT, Gemini, Perplexity and Google's AI answers cite your own site, and where a platform copy or another site gets the credit instead.