The role of PR in AI search and why it has become a core driver of visibility
- AI search visibility has changed from ranking to a process of selection, citation and recommendation
- Digital PR has become central to search visibility as credible and relevant earned coverage is more likely to appear in AI-generated answers.
- AI models prioritise consistent, credible narratives across trusted sources, turning reputation into a measurable search factor where fragmented or negative signals can directly influence AI-generated outputs.
- Future visibility will depend on building consistent authority across earned, owned, and real-world signals while fluff and shortcuts will become entirely irrelevant
I read the other day that the verb “Googling” might become obsolete within months. Whether this is an overstatement, there’s definitely a dollop of truth behind it as search has fundamentally changed.
AI search and Large Language Models (LLMs) now step in between people and the open web, therefore visibility means being selected, referenced and recommended. Studies have consistently shown how LLMs draw on signals of authority, consistency, reputation, and citation across the web. They privilege sources that are widely referenced, clearly attributed, and last but not least associated with valuable human expertise. In a nutshell: brands that earn coverage in credible and relevant publications are more likely to appear in AI-generated answers.
The latest Reuters Institute’s Journalism, Media and Technology Trends and Predictions report shows how deeply times are a-changing. AI is already transforming how information is accessed and distilled at scale, this is (allegedly) leading to publishers’ massive decline in referral traffic with dire economic consequences. At the same time, media organisations are also trying to re-engineer themselves around authority, distinctive reporting, and human expertise in order to remain visible within these new systems.
Because AI models rely on signals drawn from trusted sources to determine what is credible enough to surface, leveraging the credibility and expertise of journalists and publishers, brand visibility is increasingly reliant on earned authority and this is where the lines separating SEO and PR begin to fade. Brand visibility is no longer defined solely by the rankings, but by how reputation is encoded across the web in patterns machines can recognise and recommend.
That is why Digital PR found itself right at the centre of the fight for AI-visibility.
How did earned media become the strongest AI-visibility signal?
So, what factors actually determine which brands are picked up in AI-generated answers? Data suggests it is not alla’bout the backlinks anymore. A 2025 Ahrefs study found that brand mentions are roughly three times more closely associated with inclusion in AI Overviews than the ‘traditional SEO links’.
Backlinks still matter of course, but not as an end in themselves, which is by far the biggest change in the relationship between SEO and PR. The same Ahrefs study does suggest that backlinks might function more as proxies for authority and trust: sites with over 32,000 referring domains are 3.5 times more likely to be cited by ChatGPT.
The crucial point here is that domains with very high trust scores earn nearly four times more citations than those with lower trust scores (Ahrefs, 2025), suggesting that the common thread here is not the overall link volume, but the accumulated indicators of credibility.
More ‘traditional’ search signals such as web traffic still remain a prerequisite, as domains with 190,000 average monthly visitors receive almost double the citations of sites with lower traffic, and pages ranking between positions 1 to 45 attract significantly more AI references than lower-ranking results.
Concentrated authority is also an important concept when it comes to what moves the needle: strong organic traffic to the homepage, rather than dispersed visits to individual pages, doubles the likelihood of citation according to a research by SE Ranking, 2025.
What emerged from many of these recent studies is not a replacement of SEO fundamentals, but instead a layering of markers that have ‘authority’ as their foundation. Recognition as a reinforcement, and reputation as the deciding factor.
To PR is human, to forgive AI.
So, if earned media and brand signals help determine who is picked up, the quality of those signals matters just as much. 2026 may well be the year the old marketing adage “there is no such thing as bad publicity” is finally put to bed. LLMs are surprisingly good at cutting through discordant signals from multiple sources, connecting patterns, and distilling context – even if it means hallucinating facts to fill in any gap.
This means that any negative coverage, lack of coverage, or any inconsistent messaging can impact the narrative an AI system constructs about your brand.
Ask yourself: is your most authoritative mention from the past ten years still that Insider piece about how your company might have been in financial distress? Then, this is the time to address it.
For SEO professionals, this may initially feel unfamiliar, as a new, holistic kind of PR reputation management becomes integral to search visibility. In reality, SEOs already are in a uniquely powerful position as few functions have the strategic advantage of sitting across local search visibility, owned content, as well as PR in the way SEO does.
Strategic alignment across these channels is now essential, with local SEO responding (truthfully) to reviews and addressing issues on the ground rather than leaving unresolved criticism to linger around for AI to digest it.
Owned media must be able to tackle, acknowledge and correct your existing brand reputation where necessary instead of publishing tone-deaf fluff pieces that will only appease ‘the brand’. Alongside all this, a robust Digital PR programme is needed to secure fresh, authoritative coverage that reinforces the public narrative.
Digital PR as signal engineering and the open ecosystem
The divide between traditional PR and search-driven Digital PR is no longer very useful. Not to oversimplify, but historically, the former focused on media relationships and narratives, the latter on links and rankings, however in an AI-mediated environment, both are working towards the same goal, which is to generate and influence those signals machines interpret as ‘authority’.
What matters is not only where and when a story appears, but the information and data it contributes to the broader web environment. Individual placements fade. The cumulative pattern they form is what endures.
And this is how digital PR can directly influence the very data environment that LLMs train on and draw information from. Every earned mention, citation, and contextual reference contributes to the distributed profile machines use to interpret credibility and relevance. PR and Comms may think they are targeting journalists or audiences, but they are also, indirectly, optimising for the systems that synthesise and recommend information, it’s a type of brand influence that accrues through repeated, coherent signals that reinforce authority over time.
In this respect, it’s useful to think of your website as more than a session-grabbing destination, but as a reference source that AI models can draw from while composing their answers. If earned media can establish the narrative in public, then owned content can reinforce, clarify and substantiate it.
One of the most substantial transitions from ‘traditional’ SEO to AI search lies in this change of purpose: pages still need to attract traffic, but they must also resolve questions clearly, credibly, and with genuine information gain.
Stories and content built around engagement metrics alone decay faster when models favour recency, consistency and contextual relevance. Regularly updated pages that mirror visible real-world activities, and evidence of lived expertise now carry far more weight than evergreen, polished and brand-friendly fluff pieces – this is especially true when they introduce fresh data, bold expert-driven takes, or new developments rather than the ‘safer’ repackaging of what is already known.
The same principle extends beyond the website: press releases written purely for distribution algorithms and email pitches built around tired, formulaic hooks designed to trigger journalist templates, feel mechanical and increasingly perform poorly.
I’ve seen the future of search…it is Digital PR.
By 2027, Digital PR will be less of a supporting function and more of a commercial prerequisite for brands. To recap: AI-mediated discovery has become mainstream which means that visibility will increasingly depend on the signals brands accumulate across the open web. Mentions, citations, and consistent narratives across trusted media form the patterns large language models rely on when deciding who to recommend.
We’ve also established how there is no such thing as ‘neutral coverage’ in the LLM-powered searchscape. Press narratives and fragmented positioning can tank your chances of being surfaced faster than any Core Google Update circa 2011 ever could. In a system where recommendation logic weighs trust, coherence, and sentiment, a brand’s visibility is shaped by its overall pattern.
That also means the shortcuts will be unlikely to work. In an ecosystem attuned to tone, intent, and credibility, content written “for AI” rather than for humans risks eroding the very trust it wants to convey. What was once a credibility issue confined to extremely manipulative SEO tactics and manufactured authority signals has evolved into something much larger: a question about the integrity of the information ecosystem itself and our role within it.
Those that will dominate the future of AI search will be most likely those that embrace this framework early and avoid treating visibility within a quantity over quality framework. Refreshingly, what recent studies there have been all seem to point to, is that perhaps in the AI era, credibility will be a major competitive advantage and those who prioritise Digital PR strategies that build coherent and trustworthy narratives across journalism, owned media, and real-world activity will be the ones this new recommendation systems will favour.
Further Reading
Reuters Institute for the Study of Journalism (2026)
Journalism, Media and Technology Trends and Predictions 2026
https://reutersinstitute.politics.ox.ac.uk/journalism-media-and-technology-trends-and-predictions-2026
Ahrefs (2025)
Brand mentions show the strongest correlation with Google AI Overviews
https://ahrefs.com/blog/ai-overview-brand-correlation/
SE Ranking (2025)
How to optimise for ChatGPT and AI-driven search
https://seranking.com/blog/how-to-optimize-for-chatgpt/
Seer interactive (2006)
GEO Experiment: How AI highlighted the 1 bad review we got in 24 years
