Key Takeaways: 

• AI platforms are relying on trusted journalism to reduce misinformation and hallucinations, making earned media more influential than ever.
  
  • 89% of AI-generated outputs now cite earned media sources.
• AI and traditional search have converged, meaning that AI-generated summaries are often now the first thing people see when they search.
• Strategic communications must now consider these machine-facing audiences — ensuring your brand shows up in the queries that matter.
• Two key factors impact your brand's discoverability in AI outputs:
  
  • Fresh, credible earned media coverage
  • A consistent presence across authoritative domains

‍

It’s certainly no secret that public trust in the press has been on a decades-long decline. This lack of trust in traditional news outlets, the rise of misinformation, and our polarized media ecosystem have all chipped away at journalism’s authority. However, at Lot Sixteen we’ve been closely monitoring a significant return to earned media reliance as AI platforms lean on established journalism. 

‍

Even as public trust in journalism has declined, generative AI is heavily relying on established news outlets for its outputs in an effort to avoid misinformation and hallucinations. After all, despite the public cynicism, it’s hard to come up with a better source for broad, well-researched content and credibility than mainstream and trade publications.

‍

The New Alliance Between AI and Traditional Media
‍

In 2023, OpenAI struck its first licensing agreement with the Associated Press (AP) marking the beginning of a new era in how generative AI systems partner with outlets for reliable information. OpenAI now has partnerships with more than 20 news publishers, encompassing over 160 publications. In practical terms for the average user, when you ask ChatGPT for current analysis or news, these licensing agreements allow the model to cite a precise line or excerpt from the AP, Reuters or Washington Post and include a direct link to the original story.
‍

And it’s not just OpenAI. A recent report from Muck Rack analyzed prompts across ChatGPT, Claude, and Gemini, and found that 49 percent of citations from recent queries originate from journalistic content and a whopping 89 percent of outputs from queries came from earned media sources.
‍

As AI reshapes how consumers and businesses discover and learn about companies, the importance of third-party credibility is growing rapidly – and it’s not just limited to news outlets. Third party sources such as industry analyst content, corporate awards and rankings, think tanks, .gov, and .edu domains are also relied on heavily. 

‍

AI and Search are Converging
‍

To underscore the user behavioral shift towards AI-powered search, as of June of this year, these systems captured 5.6% of U.S. desktop search traffic, more than doubling share in just one year. This may seem like a modest figure today but the exponential growth signals the importance of getting ahead of this trend. Arguably more significant, Google has made AI overviews the default in Search, automatically surfacing generative summaries above results for most queries.

‍

There used to be a clear distinction between search engines and large language models (LLMs). At a basic level, a standard search engine was designed to be a librarian: you'd type a query, and it would return a list of links. Early LLMs, on the other hand, were like encyclopedic experts. They could answer questions directly and conversationally because their knowledge was built on vast, static training datasets. However, they didn’t access the real-time web, which meant their information had a cutoff date and couldn’t answer questions about recent news or live events.

‍

Today, this distinction is largely gone: the two technologies have converged, with AI now integrated directly into search, creating a blended approach. For instance, when you see a Google AI Overview, the system is performing a multi-step process that combines the best of both worlds: It performs a live search, retrieves relevant and authoritative web pages, and then uses a generative AI model to synthesize a summary. This allows it to combine the strengths of both systems: the real-time, live data from search indexes with the LLM's ability to present information in a natural, conversational way. Similarly, LLMs have  retrieval capabilities, enabling them to access up-to-date information from the web rather than relying solely on static training data.

‍

For brands, this means that your discoverability largely rests on two key things:

• Freshness and Domain Authority: For real-time queries, your brand's recent, credible earned media coverage is critical. These systems pull "live" information and prioritize sources with high domain authority, such as top-tier news outlets and industry-specific trade publications.
• Reputation and Patterns: For more general, non-time-sensitive queries, LLMs still rely on the "memory" built from their training data. Being consistently mentioned across multiple high-credibility outlets makes your brand a durable part of the model's understanding of your industry.

‍

Results from AI-driven search and querying shouldn’t be an afterthought for any brand at this point — in fact, it’s the new baseline. And whether someone is reading a Google AI Overview or asking ChatGPT about industry leaders in your sector, these underlying systems rely heavily on what trusted, independent sources say about your brand.

‍

AI Systems are Becoming the Front Pages of the Internet — Plan Accordingly 

‍

AI systems are fast becoming the front pages of the internet, albeit highly customized to the users queries and searches. Therefore every quote, ranking, and article in credible news outlets is not just for human audiences, it's for machine-facing visibility. This determines whether your brand will appear in the outputs that AI then delivers to real people or not. It’s important to think about the kinds of queries that your brand needs to be featured in and pursue an earned media strategy that delivers those results.  

‍

Next Time: Optimizing Owned Channels for Discoverability

‍

Earned media may be the primary backbone of credibility in AI search, but it’s not the whole story. Owned channels — from corporate blogs and research reports to FAQs and resource hubs — also make up a significant share of the links and citations that generative AI surfaces. For example, FAQ, question and response content and clear presentations of data in tables, all provide content that’s structured in a way that makes it easy for AI to lift. In an upcoming blog post, we’ll explore how to structure and optimize those owned assets so that they feature in AI outputs.