PR and AI Search Trends to Plan for in 2027

The PR and AI search trends that matter for 2027 come down to one shift: buyers increasingly meet brands through AI answers, agents, and creators rather than through a click on a publisher’s page. Plan for answer engines that cite earned media, agents that act on a buyer’s behalf, labeled ads appearing next to AI answers, and newsrooms that behave more like creators. Budget for evidence, consistency, and measurement.

Below are the seven trends we are planning around for clients in 2027, the data behind each, and what to do about them. Where a trend is a forecast, we say so. Where it is already measurable, we cite the measurement.

Photo: Megan Lee on Unsplash.

What are the PR and AI search trends to plan for in 2027?

Seven trends, ranked by how directly they affect a communications budget.

#	Trend	Signal already visible in 2026	What to do in 2027
1	AI answers become the default first impression	Google AI Mode passed 1 billion monthly users; 44% of US adults use ChatGPT	Audit and track how engines describe you every quarter
2	Earned media stays the citation backbone	84% of AI citations are earned media (Muck Rack)	Prioritize coverage in outlets engines cite for your category
3	Trend and category stories carry more weight	Trend questions draw journalism citations at more than twice the rate of how-to	Build a point of view about where your category is going
4	Agents start shopping and shortlisting	Google’s Universal Commerce Protocol and Universal Cart; OpenAI’s discovery feeds	Make product and company facts machine-readable and consistent
5	Ads arrive beside AI answers	ChatGPT ads testing; Google Direct Offers in AI Mode	Separate paid tests from organic visibility goals
6	Newsrooms act more like creators	76% of media leaders plan creator-style journalism (Reuters Institute)	Prepare executives for video, audio, and creator interviews
7	Measurement moves to share of answer	Merchant Center now offers AI share of voice insights	Add AI share of voice and description accuracy to KPIs

Will AI answers become the first impression for most brands?

For many categories, they already are. The scale is no longer in question. Alphabet said in its Q2 2026 earnings remarks that AI Mode had passed 1 billion monthly active users and the Gemini app had 950 million. Pew Research Center found 44% of US adults use ChatGPT, rising to 61% among adults under 30.

News use is earlier in the curve. The Reuters Institute’s Digital News Report 2026 found 10% of people use AI chatbots for news weekly, up from 7%, with stronger use among under-35s.

The implication for 2027 is simple: the description an engine gives of your company is now part of your reputation. Treat it like coverage. Our guide to running an AI brand reputation audit sets out how to baseline it.

Will earned media still drive AI citations in 2027?

All the evidence so far says yes. Muck Rack’s May 2026 analysis of more than 25 million cited links found earned media made up 84% of citations across ChatGPT, Claude, and Gemini, with paid or advertorial content at 0.3%. Across editions since July 2025, the earned share stayed between 82% and 89%.

The same research shows the engines behave differently:

ChatGPT cited sources in 96% of responses, averaging 5 citations.
Gemini cited in 82% of responses, averaging 8.
Claude cited in 55% of responses, averaging 13.
Top cited domains differed: Wikipedia for ChatGPT, PubMed Central for Claude, and Reddit for Gemini.

Fewer citations per answer, as in ChatGPT, means each slot is more competitive. A brand cannot rely on one engine’s habits. It needs coverage in a spread of credible sources: national business outlets, trade publications, reference sources, and the communities its buyers use. Our media directory maps how outlets such as TechCrunch and Fast Company cover different beats, and our explainer on entity strategy for GEO covers the reference layer. If the terms are new, start with our glossary entry on generative engine optimization.

Why will category and trend stories matter more?

Because engines reach for journalism when people ask where an industry is going. Muck Rack found industry trend questions drove journalism citations at more than twice the rate of how-to questions, and press releases appeared 3.5 times as often in trend answers as in best-of answers.

That is a direct signal for 2027 planning. A company that publishes credible views and data on where its category is heading gives reporters something to cover and gives engines something to cite when buyers ask the big questions. Our guide to category creation PR covers how to build those stories.

How will AI agents change discovery and buying?

Agents will increasingly shortlist, compare, and, in some categories, complete purchases. The infrastructure shipped in 2026.

Google announced the Universal Commerce Protocol in January 2026 with Shopify, Etsy, Wayfair, Target, and Walmart, and in May added Universal Cart and more UCP features across Search, Gemini, and other surfaces. OpenAI moved from in-chat checkout toward product discovery feeds in March 2026. Gartner predicted in January 2026 that 60% of brands will use agentic AI to deliver streamlined one-to-one interactions by 2028.

Agents do not read a hero banner. They read facts: what you sell, for whom, at what price, with what evidence. Inconsistent facts across your site, profiles, and coverage make you harder to shortlist. Consumer brands should read our guide on how brands get recommended by AI shopping assistants. B2B companies should treat analyst pages, review sites, and comparison content as agent inputs,, because agents read them too.

What does advertising inside AI answers mean for PR?

It creates a paid lane next to the organic answer, not a way to buy the answer. OpenAI began testing labeled ads in ChatGPT for US Free and Go users in early 2026 and has said ads do not influence answers. Google’s Direct Offers pilot shows promotions inside AI Mode to shoppers close to buying.

For 2027, keep the two goals separate:

Organic goal: be named and described accurately when buyers ask a neutral question. Earned media, reviews, and entity work drive this.
Paid goal: capture demand from people already in market. Test AI ad formats with clear conversion metrics.

Mixing the two leads to the worst outcome: paying for visibility while the organic answer still recommends a competitor.

How will newsrooms and creators change media relations?

Newsrooms will keep moving toward video, audio, and creator-style formats, and independent creators will keep taking share of attention. That changes what a good spokesperson looks like.

The Reuters Institute’s Trends and Predictions 2026 survey of 280 media leaders found YouTube was the top platform priority (net +74), with AI chatbot distribution second (+61). Seventy-six percent planned to have journalists behave more like creators and 50% planned to partner with creators. Leaders also expected search traffic to their sites to fall by more than 40% over three years.

What this means in practice:

Executives need to be good on camera and on audio, not only in print.
Independent journalists, newsletter writers, and creators belong on the media list.
Outlets losing search referrals will value exclusives, data, and access more than generic news.

What do most teams get wrong about AI search planning?

They plan for one engine, one tactic, or one quarter. The most common mistakes we see going into 2027:

Chasing a single platform’s quirks. Engines cite differently. Plan for a spread of credible sources, not one engine’s favorite domain.
Expecting technical fixes to do earned media’s job. Structured data and llms.txt help engines read you. They do not create independent evidence.
Measuring only clicks. Fewer people click through from AI answers. Measure inclusion, description accuracy, and branded demand as well.
Letting each team own a fragment. PR owns coverage, SEO owns the site, ecommerce owns feeds, and nobody owns the answer. Name one owner for AI visibility.
Planning annually. AI answers shift as models and retrieval change. Review quarterly.

How should a CMO build a 2027 plan around these trends?

Use a five-step planning sequence before budgets lock.

Baseline AI answers. Run a fixed set of 40 to 60 prompts across ChatGPT, Gemini, Claude, Perplexity, and Google AI Mode. Record inclusion, description, and cited sources.
Choose the category story. Decide the point of view you want engines and reporters to associate with you, and the category term you will use everywhere.
Map the sources engines cite. List the outlets, reference pages, review sites, and communities that appear in your category’s answers. That list becomes your target media list.
Build the evidence calendar. Plan original research, executive commentary, product news, and customer proof by quarter, each aimed at specific sources on the map.
Set AI visibility KPIs. Add AI share of voice, description accuracy, and cited-source mix to the PR scorecard, next to coverage quality and branded search. Our guide to AI share of voice explains how.

What could this look like for a growth-stage company? A hypothetical example

This is a hypothetical scenario for illustration, not a client story.

A Series B security software company runs its baseline in December. ChatGPT and Gemini describe it accurately but leave it out of “leading vendors” answers. The cited sources are two analyst-style comparison pages, a trade outlet the company has never pitched, and a Reddit thread.

Its 2027 plan follows the sequence. The company chooses one category story about identity attacks on small IT teams and commissions a survey to support it. It pitches the trade outlet first, offers the CEO for a video interview series run by a newsroom that has shifted to creator-style formats, and corrects its own comparison page to match how buyers phrase the question. It budgets a small paid test for AI ad formats, measured separately.

The quarterly re-run is the scorecard. The goal is not a guaranteed placement in any answer. It is a steady increase in inclusion and accuracy across engines, supported by sources the company can point to.

Frequently asked questions

What is the biggest AI search trend for PR in 2027?

The biggest trend is that AI answers are becoming the first impression of a brand, and those answers rely mostly on earned media. Google AI Mode passed 1 billion monthly users in 2026, and Muck Rack found 84% of AI citations come from earned media. PR teams should treat AI descriptions as part of reputation and measure them.

Will press releases matter in 2027?

Press releases still matter as a factual record, especially for trend and industry questions, where Muck Rack found they appear 3.5 times as often as in best-of answers. They work best as a citable source on your newsroom, supporting a story that reporters cover independently, not as a substitute for pitching.

Should brands advertise inside ChatGPT or Google AI Mode?

Test it with a small, separately measured budget if your buyers are in market there. Ads in ChatGPT and Direct Offers in AI Mode are labeled and sit alongside answers. They do not change the organic recommendation, which still depends on evidence such as coverage and reviews.

How often should we check how AI engines describe our brand?

Quarterly at minimum, and monthly during launches, funding announcements, or crises. Engines that retrieve live information can change their answers within weeks, and model updates can change answers without warning. A fixed prompt set makes results comparable over time.

Do creators count as media for PR in 2027?

Yes, when they have editorial standards and an audience your buyers trust. Media leaders surveyed by the Reuters Institute planned to partner with creators and have journalists behave more like creators. Disclose any paid relationships, and treat creator interviews with the same preparation as a press interview.

If you want a 2027 plan that connects earned media to how AI engines describe your company, our AEO, GEO, and search visibility team and media relations team work as one program. For the year’s backdrop, read our 2026 PR year in review, then talk to us about a baseline.