Want Your Brand to Show in ChatGPT Results? Time to Review Your PR Program

“I Quit Google Search for AI—and I’m Not Going Back” 

Headlines like this, which appeared in The Wall Street Journal last month, reflect a notable change in user habits: Many of us are turning to AI-powered tools like ChatGPT, Gemini, Grok and Perplexity to look up things we would have previously searched in Google. (It’s even spawned a new marketing acronym: GEO or Generative Engine Optimization).  

While I’m not ready to shelve “Googling” as a verb, I do agree with many of my peers that it’s time to consider what the move to large language models (LLMs) means for marketers. What happens to search engine optimization (SEO) as users begin to shift away from “search engines”? What happens to PR and media relations when readers start their journey on ChatGPT and not The Wall Street Journal? 

I have an answer – and it might surprise you: With the increasing prominence of generative AI and LLM, SEO and PR are more important than ever.  

In this blog we explore how generative AI and LLMs are using earned media coverage and what media publications tend to be indexed most frequently.  

How LLMs are Using Earned Media Coverage 

At Mod Op, we’ve launched a multidisciplinary team of PR professionals, SEO specialists and members of our AI and innovation team to explore, understand, and use AI tools to strategize improved brand presence in AI-generated content.  

One of our early findings: ChatGPT tends to cite well-known, authoritative media sources. In other words, mainstream news sites with strong reputations and visibility, like The New York Times (NYT), CNN and Forbes, are heavily represented in ChatGPT’s citations.  

How do brands secure mentions in publications like The New York Times (NYT), CNN and Forbes? By effective use of PR and, more specifically, media relations. In fact, MOZ founder Rand Fiskin even cites LLMs and AI indexing as one of the reasons that public relations is the future of marketing:

“Language models are governed by the ‘trusted sources’ they crawl, index, and build up their text corpora,” explains Rand, “Three guesses what influences change in those models. Yup, it’s PR-Influenced Content. The ability to influence how people write about, talk about, and publish about you on the web directly impacts how AI tools respond to questions about your brand, your field, and whether they include you when prompters ask about the problems you solve.” 

What Media Outlets is AI Indexing? 

To better understand what media outlets are being indexed by LLMs, we decided to look at one of the most popular: ChatGPT. We used ChatGPT4.5 + Deep Research to complete an analysis of what publications are cited most often in ChatGPT, focusing on tech and consumer goods.  

Major Media Outlets 

According to our analysis, major U.S. outlets like The New York Times (NYT), The Wall Street Journal (WSJ), and CNN frequently appear as cited sources due to their reputation and high search visibility (DeepSeek vs. ChatGPT vs. AI Overviews: YMYL Research Study).   

In PR — as well as SEO — we often use metrics like UVM (unique visits per month), which estimates an outlet’s viewership (the modern “circulation”), Domain Authority and Spam Score to judge an outlet’s reputation and search visibility.   

News agencies such as Reuters are also commonly referenced (DeepSeek vs. ChatGPT vs. AI Overviews: YMYL Research Study). Like the major U.S. outlets mentioned above, Reuters has a relatively high UVM (50.6k), very strong Domain Authority (94) and low spam score (3%). 

Industry-Specific Publications 

Alongside general news, ChatGPT cites industry or sector-focused journals and trade publications when addressing industry-specific queries.  

In the technology sector, leading tech news sites like TechCrunch, Wired, The Verge, Engadget, and CNET are frequently referenced (Top 100 Tech News Websites for Technology Enthusiasts in 2025). These outlets specialize in tech coverage and often rank highly for tech queries. Again, if we look at common PR and SEO metrics, we see a pattern of relatively high UVM, very strong Domain Authority and low spam score. 

For consumer goods and retail topics, trade publications such as Retail Dive and Consumer Goods Technology are often cited for industry trends and analysis 

(Retail Dive: Retail News and Trends) (Consumer Goods Technology: Consumer Goods Industry News …). Similarly, Consumer Reports (for product reviews) and Chain Store Age (for retail news) may appear in citations due to their domain expertise, according to our ChatGPT4.5 + Deep Research analysis.  

For these more niche publications, UVM becomes less clearly indicative of reputation. We also begin to see Domain Authority dip below excellent, but it’s within the range of what is considered “good” (50 – 60). The spam score remains low.   

Wikipedia  

One interesting source that emerged from our ChatGPT4.5 + Deep Research analysis: Wikipedia. While not a media outlet, Wikipedia is one of the most cited sources across many topics (DeepSeek vs. ChatGPT vs. AI Overviews: YMYL Research Study). This is particularly interesting when you consider the connection between Wikipedia and earned media coverage. 

If you’ve ever tried to make a change on Wikipedia, only to have it quickly scrapped by Wikipedia editors, you’ve likely already discovered the importance of sources.  

Earned media coverage acts as a powerful citation engine for Wikipedia pages. When authoritative media outlets write about a topic, person or an organization, these articles serve as independent, verifiable sources that Wikipedia editors can use to support claims.  

Publications Most-Cited by ChatGPT  

Developed using ChatGPT4.5 + Deep Research, the list below ranks some of the most-cited publications in ChatGPT’s outputs for tech and consumer goods topics (U.S. focus), based on their observed citation frequency in ChatGPT’s answers. 

1. Reuters 
2. The New York Times 
3. CNN 
4. Forbes 
5. TechCrunch 
6. Wired 
7. The Verge 
8. CNET 
9. Retail Dive 

While it’s important to note that ChatGPT4.5 – even with DeepResearch – and make mistakes, this is a great starting point.  

As you’re looking to PR and media relations to further brand awareness, and how you may be able to use earned media to support AI indexing and AI search visibility, I recommend using a media database like MuckRack to review the UVM, Domain Authority and Spam Score for various target publications. After all, LLMs index publications with smaller UVMs and lower Domain Authority, especially when it’s around niche topics — for example, Consumer Goods Technology, with a UVM of under 100k and Domain Authority in the “good” range (50 – 60).   

With all that in mind, here are ten publications I’d recommend exploring: 

Some final thoughts: consumer search behavior is changing. With users shifting from search engines to AI-powered tools like ChatGPT and Gemini, the role of earned media has never been more critical. For those with established PR programs, refining media targets by integrating metrics like UVM, Domain Authority, and Spam Score can improve the quality and credibility of earned media placements, increasing the likelihood coverage will be cited by AI models like ChatGPT. For those without a PR program – it’s time to start building.