AI and the Future of PR and Digital PR
PR has quietly become one of the most important AI-marketing disciplines — because LLM citations are decided largely by who PR put on which sources. A 2026 UK discipline report.
PR was, until very recently, the marketing discipline most likely to be cut when budgets tightened. Hard to measure, slow to deliver, expensive relative to performance channels — the strategic case for PR had been eroding for the better part of a decade. That trend has reversed sharply in 2026, and the reason is genuinely structural: AI engines have become a major mediator of what brands customers consider, and AI engines decide what to recommend largely based on what credible sources have already said.
That has put PR — earned coverage in trusted publications, named expert positioning, original-research placement — back at the centre of marketing strategy. The brands that maintained PR investment through the lean years are now finding it disproportionately valuable. The brands that cut it are now trying to rebuild it from a standing start, with predictably mixed results.
The state of play in 2026
- Volume PR. Wire-service press releases, broad distribution, low-substance outreach. Once a meaningful share of UK PR activity, now producing minimal measurable value. This model is in irreversible decline.
- Targeted earned PR. Story-led pitching to specific journalists with genuine substance. Always the gold standard, now more valuable than ever as the only model that produces both traditional reach and LLM citation lift.
- Earned discoverability (the new discipline). A broader practice that includes targeted earned PR plus deliberate presence work on LLM-citation sources, reputation auditing across AI engines, and named-author placement strategy. A small but growing set of UK PR practices have repositioned around this scope.
Average increase in LLM citation likelihood observed for UK brands receiving Tier-1 UK publication coverage versus equivalent brands without — measured as named-brand mentions in ChatGPT, Perplexity, and Claude responses to relevant commercial-intent queries.
What's actually changing
- The sources that matter have narrowed. LLMs cite disproportionately from a relatively small set of high-trust sources. Coverage on those sources is now meaningfully more valuable than coverage on equivalent-traffic sources outside that set.
- Original substance is the unit of trade. Journalists and AI engines both reward proprietary data, named expert thinking, and credible original perspective. PR with nothing substantive underneath produces nothing of value either side.
- Spokesperson strategy has become structural. Named senior leaders with consistent expertise positioning earn disproportionate coverage and citation. PR practices winning in 2026 invest meaningfully in building specific people as recognisable sources, not just placing logos.
- Reputation is now multi-surface. What Google says, what ChatGPT says, what Perplexity says, what Reddit says — all matter, all are auditable, and all are addressable through different PR-style interventions.
Live UK signals
- UK Tier-1 publications (FT, The Times, BBC, The Guardian, City AM, sector-leading trade press) have seen meaningful increase in unsolicited brand-pitch volume from AI-aware brands seeking placement explicitly for LLM citation purposes.
- Reddit's role as an LLM training and citation source has driven a small wave of UK brands experimenting with deliberate Reddit community presence — with mixed and brand-dependent results.
- Original-research placement is the fastest-growing UK PR sub-discipline in 2026, driven by the convergence of editorial appetite for proprietary data and the LLM citation premium for the same content.
- Reputation monitoring across AI engines has emerged as a paid service category, with several UK practices now offering quarterly AI-reputation audits as a productised service.
What this means for UK CMOs
- Reinvest in PR — but reframe the function. If your PR budget was deprioritised over the last five years, 2026 is the moment to rebuild it, but not as the function it was. The brief is earned discoverability, not just press coverage.
- Build substance before booking PR support. Original research, founder positioning, distinct point of view. PR multiplies what is there; it cannot create it.
- Audit your AI reputation quarterly. Knowing what ChatGPT, Perplexity, Claude, and Gemini say about you is now a baseline competence, not an advanced tactic.
- Identify two or three named voices to invest in. Spokesperson strategy compounds. Pick the people, commit to the cadence, and stay with it.
For the broader context, see SEO in the GEO Era and AI and the Future of Content Marketing.
Frequently asked
Is PR really more important now than five years ago?
Yes, and the data supports it. The combination of AI-mediated discovery (LLMs choosing what to recommend), declining trust in branded content, and the increasing role of named-author signal in both SEO and AI citation has made earned coverage from credible sources structurally more valuable than at any point in the last decade. Multiple UK brands report that a single Tier 1 placement now drives more measurable downstream value than it did in 2020.
What is 'PR for LLMs'?
It is the practice of placing brand mentions, expert quotes, and proprietary data on the sources that major LLMs cite from when answering relevant queries. Those sources are knowable: established UK news publications, Reddit (in many categories), Wikipedia (where genuinely warranted), specific industry analysts and trade publications. Coverage on those sources increases the likelihood of being mentioned by AI engines on related queries — a measurable, growing channel.
Are press releases still useful?
Largely no, in the volume-distribution sense they have been used for the last twenty years. Wire services and undifferentiated press release distribution produce minimal earned coverage and almost no LLM citation lift. What does work in 2026 is targeted, story-led outreach to specific journalists with genuinely original material — exactly the same thing that has always worked, with even more emphasis on substance because the alternative (low-effort coverage of low-effort releases) has commoditised toward zero value.
How is AI changing PR operations themselves?
AI is being used heavily for media monitoring, journalist research, sentiment analysis, briefing preparation, and first-draft pitch writing. The PR teams operating well are using AI to multiply their reach into press lists they previously could not service, while keeping the human relationship-building at the centre. Pitching itself is increasingly AI-assisted but cannot be AI-led — journalists detect and discount obviously AI-written outreach immediately.
What about reputation management when AI is the audience?
Brand reputation in 2026 increasingly includes how AI engines describe you when asked. Auditing what ChatGPT, Perplexity, Claude, and Gemini say about your brand on relevant queries should be a quarterly minimum for any UK brand of meaningful size. Negative or inaccurate AI descriptions can sometimes be addressed through PR-style corrective work on the source content the LLMs are drawing from. This is a new sub-discipline and will continue to professionalise rapidly.
How should a UK CMO actually invest PR budget in 2026?
Reframe the function as 'earned discoverability', not just 'press coverage'. Allocate roughly: 40% to substance production (original research, founder thinking, customer evidence) — without it, no PR budget produces results, 35% to targeted media and analyst placement, 15% to LLM-citation-source presence work (Reddit, Wikipedia, specific high-trust publications), 10% to monitoring and reputation auditing across both traditional media and AI engines. The shape is different from the 2020 PR retainer for a reason.
Book a strategy call with Kasim Javed
Bring this thinking to your business. Spend 30 minutes with Kasim — CEO of Creative Marketing Group — to map where AI can absorb the most cost and unlock the most growth in your marketing operation. Or skip the diary and pressure-test your strategy live with KJ-AI, his always-on Chief Marketing Officer agent, right now.
Related insights
All researchThe End of the Marketing Funnel
How AI agents and zero-click search collapse the classic funnel — and the new shape of demand that replaces it.
AI and the Future of Marketing Analytics
Attribution is broken, dashboards are dying. A 2026 UK discipline report on the causal, agent-driven measurement stack that replaces them.
SEO in the GEO Era
Classic SEO has split. A 2026 UK discipline report on Generative Engine Optimisation and the rebuild of organic discovery for the AI answer-engine era.