From Search to trust: Bridging the literacy gap with GEO
TL;DR
- LLMs are reshaping how people begin health journeys.
- AI can aid literacy but risks worsening inequalities.
- Trust now depends on visible, verifiable credibility.
- Earned media drives AI visibility and trust.
- GEO supports equity where health conversations start
In this era of LLMs, the intersection of public relations (PR) and artificial intelligence is reshaping how healthcare brands connect with audiences.
People no longer look for information the way they did even a few years ago. Instead of asking Jeeves, they’re having real-time conversations about their concerns, curiosities and questions. Large language models (LLMs) such as ChatGPT, Claude and Gemini (among others) are reshaping how we find information – and how we understand it.
The active number of daily users interacting with LLMs grew by 233% between November 2024 and April 2025 as a result of increased awareness and the deeper integration into daily tasks [1]. The rapid rise of generative AI now extends beyond everyday questions. People now see these tools as personalised navigators for their health journey, turning to them when they feel anxious, need urgent help, want to make sense of symptoms, or prepare for appointments.
These tools offer conversational and human-like answers, taking complex information and distilling it into something that’s easy to make sense of. This saves people from the mental exhaustion of trying to understand what they’re reading and what it means for them.
In January, OpenAI introduced ChatGPT Health, seeing an opportunity after 2025 data revealed that ChatGPT received 230 million health questions each week. Currently only rolled out in the United States and outside of the European Economic Area, Switzerland, and the UK, ChatGPT Health offers a dedicated and protected space designed for health‑related conversations. [4]
This shift represents a fundamental change in how health and pharmaceutical brands can build trust and support patient understanding before a clinician comes into the picture.
Why does this matter?
Health information can be stressful. In moments that can otherwise feel lonely, a conversational response can feel more approachable than a wall of links. LLMs resolve issues end to end, giving patients clear responses, guidance on what to do, and reassurance about their health. But we shouldn’t see LLMs as a cure-all for patient woes. Digital literacy matters, and without caution AI could widen existing health inequalities. With more people using AI as their first port-of-call in their health journey, it’s up to us as health communicators to understand how these tools shape patient beliefs and decisions. Ignoring this shift in search risks misinformation filling in the gaps and patients not adhering to the advice they receive from HCPs. This reality can be particularly troublesome for those with lower literacy levels or who don’t use English as a first language, as they may be even more reliant on AI’s ease of communication.
This creates both opportunity and risk: while AI can support better health literacy, there is still a gap in trusted, authoritative voices.
If health communicators do not take part in GEO (Generative Engine Optimisation), misinformation will fill the gap. As generative tools become part of everyday health research, there is a clear responsibility to provide accurate, contextual and patient‑centred information. [3]. Putting the patient at the centre of GEO supports safety, trust, equity and health literacy. It also recognises that many patient journeys now begin with a conversation with a digital tool rather than a visit to a waiting room.
Trust is key
Following a Guardian investigation, it was revealed that Google had removed some of its AI-generated summaries for certain health searches, which had been described as ‘dangerous’ by experts.
Trust is central to health communication. How people build trust has changed greatly. Today, people expect not just accuracy, but credibility that they can see and verify, often in real time.
This makes visibility essential. In practice, this means that content, insights, and research need to be discoverable where people are already searching. This includes AI-powered discovery platforms, social media, or traditional news outlets. We’ve seen clients’ news and thought pieces appear on Perplexity Discover page, making information accessible in ways that were not possible prior. This kind of exposure does not just push content out into new channels. It reinforces trust by showing that third party sources recognise and cite your work.
By integrating news, research citations, and community-focused messaging, organisations can strengthen both visibility and trust simultaneously, creating a cycle that supports ongoing health communication goals.
The Power of earned media in AI-driven communications
Earned content (mentions and coverage that brands have not paid for) has long been a cornerstone of effective PR. Unlike advertising, where you control the message and purchase space, earned media relies on others saying great things about you, creating independent and credible narratives. With the rise of LLMs, this principle has gained new significance.
Recent data shows that nearly 95% of links cited by AI are non-paid, with 89% coming from editorial articles in newspapers and magazines. [5]
This means that AI-driven platforms increasingly favour content that is authentic, topical, and relevant, making PR and communications critical pathways to visibility. Brands cannot simply pay to influence LLMs; instead, they must earn attention through strategic storytelling and engagement.
Understanding how LLMs source and cite information
Not all LLMs operate identically. Different models prioritise different sources, and their citation practices are continually evolving. For example, OpenAI recently reduced its reliance on Wikipedia, while Gemini has become more likely to cite YouTube content. Press releases are seeing a rise in citations, and 51% of AI-generated answers include a reference to Reddit. Fresh content is favoured, with material published in the last 12 months more likely to be featured. [5]
This shows that trust is built through earned presence in the form of community recognition, citations, expert commentary, and thought leadership. Over time, users become accustomed to this cumulative visibility and develop trust.
However, reports are showing that there is a disproportionate use by LLMs of some outlets as some media, such as the BBC, has sought to block AI companies from accessing content. ChatGPT seems to be respecting this, whereas Gemini and Perplexity are not. The Institute for Public Policy Research (IPPR) highlights that different licensing agreements impact which sources are cited, and makes recommendations for ensuring that AI improves the quality and diversity of information that people access. [6]
The rise of AI is also threatening the financial sustainability of quality journalism. Independent journalists play a vital role in holding power to account, challenging misinformation and helping people make sense of complex issues. The risk is that AI is drawing from news outlets, at the same time as impacting their ability to provide credible, relevant content.
A recent example of the threat faced by media outlets, is the news in early February that The Washington Post will lay off approximately 300 journalists from across sports, culture, local staff and journalists working in locations such as Ukraine and the Middle East. A statement issued by the Post said: “The Washington Post is taking a number of difficult but decisive actions today for our future, in what amounts to a significant restructuring across the company. These steps are designed to strengthen our footing and sharpen our focus on delivering the distinctive journalism that sets The Post apart and, most importantly, engages our customers.” Critics of the decision have focused on The Post’s owner, Jeff Bezos, and highlighted the issue of journalism under threat. Marcus Brauchli, the Post’s executive editor until 2012, said: “These are historic times, given the cyclone bearing down on the world order and American system of government. This is when journalism matters most. I mean, laying off reporters in Ukraine, now.”
Optimising PR for LLMs: Key strategies
Optimising for LLMs requires a distinct approach from traditional earned media strategies. The structure and composition of press releases matter more than ever. Including statistics, bullet points, and clear grammar can increase the chances of being cited by AI systems. Additionally, the journalists and outlets most influential for LLM citations may differ from those traditionally targeted in health and corporate life sciences communications.
For non-press release content, conversational copy and direct answers to common questions improve the likelihood of citation. The process is not a quick win – earned media takes time and effort. The value lies in independent, credible, and relevant stories that stand out in the AI-powered information ecosystem.
Engaging thoughtfully allows health communicators to maximise impact, reinforce trust, and actively contribute to equitable access to lifesaving information, while navigating a complex digital landscape responsibly. In this sense, GEO is a health equity lever. It is about expanding health communication to where patient understanding begins. That’s a conversation with a generative AI tool.
If you want to put earned content to work in a strategy optimised for AI visibility and trust, get in touch at st*********@***********th.io
References
- The New AI Advantage: Daily AI-Users Feel More Productive, Effective, and Satisfied at Work
- How PR can own the future of search: A tactical guide to GEO | Muck Rack Blog
- Health organisations raise concerns over AI summaries on search engines
- Introducing ChatGPT Health
- What’s AI reading?
- AI’s got news for you – Can AI improve our Information environment? | IPPR Report Jan 2026
GEO is the practice of ensuring accurate, contextual, patient‑centred information is discoverable within generative AI experiences so trusted voices show up where people are increasingly “searching.”
The blog suggests that if health communicators don’t participate, misinformation can fill the gap especially as patient journeys increasingly begin with AI tools.
Instead of browsing a “wall of links,” people are having real‑time conversations with tools like ChatGPT, Claude and Gemini to make sense of concerns and symptoms.
The blog frames LLMs as “personalised navigators” that can help people prepare for appointments or understand what they’re reading.
The blog cautions that AI isn’t a cure‑all: digital literacy matters, and without care AI could widen health inequalities and shape patient beliefs and decisions in unhelpful ways.
It also highlights emerging trust concerns, referencing reporting that Google removed some AI‑generated health summaries described as “dangerous” by experts.
Earned media (unpaid mentions and coverage) strengthens credibility because it relies on independent voices rather than paid placement.
The blog notes that AI citations skew heavily toward non‑paid links, with a high share coming from editorial articles making PR and communications a key pathway to visibility
For press releases, structure matters: the blog suggests that including statistics, bullet points and clear grammar can increase the likelihood of being cited.
For non‑press content, it recommends conversational copy and direct answers to common questions while emphasising that earned media is a long‑game, not a quick win.
