Digital PR and AI Citations, Explained: Why ChatGPT Trusts Earned Media Over Your Website

TL;DR

• What: digital PR is the work of getting a brand covered and mentioned by online media, industry sites, blogs, podcasts and experts. In AI search, it has become one of the main ways brands get cited.
• The data: in Muck Rack’s May 2026 study of more than 25 million links cited by ChatGPT, Claude and Gemini, 84% came from earned media and 27% from journalism alone.
• Why: AI search services consistently favour independent, authoritative sources over brands’ own websites and social media, according to a 2025 study comparing them with Google.
• Mentions over links: in an Ahrefs study of 75,000 brands, branded web mentions correlated with AI Overview visibility about three times more strongly than backlinks.
• What to do: find out which sources AI cites in your niche, give journalists data and stories, keep your brand description consistent, and measure monthly.

Contents

1. How AI assistants choose sources
2. What the data shows
3. Why AI trusts earned media
4. Mentions vs. backlinks
5. What digital PR includes
6. How to start
7. What doesn’t work
8. FAQ
9. Vyara’s take

More people now ask an AI assistant which company, product or service to choose before they search Google. The brands named in those answers are rarely chosen at random. They are the brands that other sources write about.

How AI assistants choose sources

AI assistants such as ChatGPT, Gemini, Claude and Perplexity draw on two kinds of information: what the model learned during training, and live web search. For questions that need current information, such as recommendations, prices, reviews or news, the assistant searches the web, selects a few sources and builds its answer from them.

Those sources shape the answer itself. In Muck Rack’s research, turning web search on or off noticeably changed the AI’s responses. For a deeper look at the mechanics, see How AI Search Decides Who to Cite.

What the data shows

Muck Rack defines earned media broadly: journalism, academic research, government sites, encyclopedias and third-party content about a brand. Press release citations fell compared with its December 2025 report, which the company attributes mainly to ChatGPT citing them less often.

Muck Rack sells PR software, so it has an interest in these findings. Its results have been consistent across three editions, and they align with independent research.

Why AI trusts earned media

A 2025 study by Mahe Chen and colleagues, published as a preprint and not yet peer-reviewed, compared AI search services with Google. AI search showed a strong preference for earned media, meaning third-party and authoritative sources, over brand-owned and social content. Google returned a more balanced mix.

The reasoning is straightforward. A brand calling itself “the leading company” on its own website is a claim. The same statement in an independent publication, a verified review or an industry award is closer to evidence. AI assistants are designed to give answers that can be verified, so independent sources carry more weight.

The type of question matters too. When people ask what is happening in an industry, AI turns to news coverage. When they ask how to do something, it relies more on reference content and brands’ own sites. Brands need both.

Mentions vs. backlinks

Traditional SEO measured authority largely through backlinks. Ahrefs’ study of 75,000 brands points elsewhere for AI search: branded web mentions had a correlation of 0.664 with visibility in Google’s AI Overviews, compared with 0.218 for the number of backlinks. Ahrefs notes that correlation is not causation.

The likely explanation is context. To recommend a brand, an AI system needs to understand what the brand is known for, treating it as a distinct entity. It learns this from the text surrounding the brand’s name. Repeated mentions in articles about a specific topic build that association. Links from directories without meaningful text do not.

What digital PR includes

Digital PR covers earning coverage and mentions in online media, industry publications, blogs, podcasts and expert commentary, with or without a link. Unlike traditional PR, its results are measurable: where a brand appeared, who mentioned it, and whether it brought traffic or enquiries. Unlike link building, the goal is the mention and its context, not the link itself.

Typical reasons for coverage include:

• original research or surveys;
• data only the brand has;
• expert commentary on a current topic;
• a story behind the brand;
• a useful tool or ranking that media want to reference.

For example, a renovation company that surveys 300 clients about the average cost of a bathroom renovation in its city has created news. If media cite the company as the source, AI systems gain a reason to associate it with that topic.

How to start

1. Map the sources. Ask ChatGPT and Gemini around ten questions real customers would ask, and record which sites are cited.
2. Give journalists something to use. Data, stories and clear positions work better than product press releases.
3. Offer expertise. Comment on topics where you are strong and build relationships with journalists covering your industry.
4. Stay consistent. Describe your company the same way in media, profiles and on your website.
5. Measure monthly. Repeat the same questions and track whether and how you appear. Meaningful change typically takes at least three months.

What doesn’t work

• Mass distribution of press releases without real news value.
• Paid content presented as editorial coverage.
• Low-quality, bought links on sites with no real audience.

FAQ

Vyara’s take

“Until recently, PR was a matter of image with people. Now it’s also a matter of image with AI. When you call yourself the best on your own site, that’s a claim. When a journalist writes it, the AI treats it as evidence. I recently worked with a brand that had left negative reviews from a smear campaign unanswered for almost seven years. People kept finding them. So did AI.”

Further reading: How to Build Digital PR So AI Recommends You – Vyara Stefcheva’s personal take on stefcheva.com.

Sources: Muck Rack, What Is AI Reading?, May 2026 · Muck Rack report, 2025 · NNA on the May 2026 report · PAGE on Muck Rack’s findings · Chen et al., arXiv 2509.08919 · Ahrefs, 75K brands study

About the Author

Vyara Stefcheva

Vyara Stefcheva is the founder and editor of Egorithms and an independent marketing expert. She follows how technology, AI and algorithms change the way people find, choose and buy, and writes about what that means for businesses and for the people who use them.

She has spent her career building marketing and business strategies for more than 300 brands across three continents. She built and exited a marketing agency and now works independently. She has taught more than 20,000 people through lectures, courses and corporate training, and speaks at conferences about AI and the future of the internet.

On her YouTube channel she tests new AI and marketing tools as they launch and gives each an independent verdict: adopt, watch or ignore.

See author's posts