Your Press Coverage Is Now LLM Training Data

The deal that settled the question was signed quietly in 2024. Google paid Reddit $60 million a year for the right to use its posts as training data. OpenAI followed shortly after with a separate licensing agreement. What those deals confirmed is something that marketers in Malaysia have only recently started to absorb: when AI systems answer a question about your industry, your competitors, or your brand, they are drawing on web content — and digital PR is now one of the most direct ways to influence what they learn.

Digital PR in Malaysia has traditionally been justified by two things: brand awareness and backlinks. Those outcomes still matter. However, a third outcome is emerging that most PR strategies here have not yet priced in — training AI to associate your brand with authority, expertise, and credibility in your market. This article explains how that works and what to do about it.

The Deal That Changed the Meaning of Press Coverage

Google’s $60 million annual deal with Reddit, confirmed by Ars Technica in May 2024, was not a licensing quirk. It was an explicit statement about how large language models get built. Reddit’s billion posts and sixteen billion comments are a gold mine for AI companies because they represent how real people describe problems, evaluate products, and discuss brands in natural language. Google paid that sum because that content shapes what AI systems know.

The OpenAI deal followed the same logic. OpenAI gained access to Reddit’s real-time Data API, giving its models the ability to incorporate current conversations — not just historical archives — into their understanding of the world. In other words, what people say about your industry on the internet is now formally, contractually, part of the AI knowledge layer.

For companies doing digital PR Malaysia, this changes the calculation. A feature in The Edge or a mention in Marketing-Interactive is no longer just a placement that raises brand awareness. It is a signal that LLMs weigh when they decide what to say about your market, your category, and your business.

Why Third-Party Coverage Outperforms Your Own Blog

Your website matters. However, AI systems are specifically designed to weight independent, third-party sources more heavily than self-published claims. The reason is straightforward: a company can write anything about itself, whereas a journalist, editor, or analyst is assumed to have done independent verification before publishing.

Research from Discovered Labs puts this dynamic plainly: one editorial placement in a trusted outlet does more for AI citation rates than ten blog posts on a standalone domain. That is because LLMs evaluate not just what your site says but how consistently that information appears across independent sources — what researchers call information consistency. When your positioning appears on your own site, in three trade publications, in Reddit discussions, and on a comparison platform, the convergence builds the consensus that LLMs trust.

The Princeton GEO research, presented at KDD 2024, quantified part of this dynamic. Content that includes direct quotations from named external sources produces a 41% lift in AI citation visibility. Content containing statistics with named sources produces a 40% lift. In contrast, keyword-heavy content with no external validation produced roughly 10% worse results than baseline — meaning classic SEO padding actively harms AI citation rates.

What this means for digital PR Malaysia practitioners is direct: the standard goal of “we got a placement” is no longer sufficient. What matters is whether the placement creates a verifiable, citable signal that LLMs can use to validate your brand’s claims.

Which Malaysian and Singaporean Outlets Actually Train AI

Not all press coverage carries equal weight in AI training datasets. Large language models pull from web corpora, but those corpora are dominated by high-authority domains that AI companies explicitly target for licensing and crawling. For the Malaysian and Singaporean market, the outlets that matter most for AI training signal fall into a clear hierarchy.

Tier 1 — High AI training signal:

The Star (Malaysia): Highest-circulation English daily in Malaysia, regularly crawled and indexed. Business, tech, and property verticals are well-represented in AI responses to Malaysian queries.
The Business Times (Singapore): Financial and corporate authority for the Singapore market. Frequently cited in AI responses to SG-specific business queries.
The Edge (Malaysia/Singapore): Business and capital markets authority. Strong signal for B2B and financial service brands.
TechInAsia: The highest-reach English-language tech publication serving Southeast Asia. Regularly cited in AI responses about SEA startups, tech adoption, and digital economy trends.

Tier 2 — Solid AI training signal:

e27: Singapore-based startup and innovation media, strong for B2B tech and SaaS brands targeting investors and founders.
KrAsia: Pan-Asian tech coverage with strong indexing in AI training corpora for Southeast Asian topics.
Marketing-Interactive: The leading B2B marketing trade publication for the SG and MY market. Brands that appear here earn credibility with the audience most likely to evaluate their services.
Marketing Magazine Malaysia: The equivalent trade authority for the Malaysian marketing sector.

A note on Bahasa Melayu outlets: Berita Harian, Sinar Harian, and Astro Awani carry strong reach but are underrepresented in the English-language training corpora that most international LLMs are built from. Their value is real for Bahasa-first search queries, but for AI visibility in English-language responses, the Tier 1 and 2 outlets above have greater leverage for now.

Digital PR Malaysia: What the Strategy Must Look Like Now

A digital PR strategy built around AI visibility looks different from a traditional media relations campaign. There are five specific adjustments that matter for the MY/SG market.

1. Lead with proprietary data

Research published by Foundation Inc’s 57-million citation study found that content containing original research earns editorial coverage at rates that generic commentary cannot match. When Malaysian brands publish their own market data — a survey of local SMEs, a benchmarking report on industry adoption — journalists have a reason to cite the source directly, rather than paraphrasing a claim. Those citations create exactly the independently-corroborated signals that LLMs weight most heavily.

2. Pitch the MY/SG-specific angle

International publications cover Malaysia and Singapore, but they cover them selectively. A pitch framed as “how this global trend plays out differently in the JB-Singapore corridor” is more likely to land in a local edition than a generic product announcement. More importantly, MY/SG-specific coverage is relatively rare — which means a strong local placement carries disproportionate weight for queries about the Malaysian and Singaporean market.

3. Maintain naming consistency across platforms

LLMs struggle to consolidate mentions into a single entity when a brand appears under different names across platforms — “Xwork,” “Xwork Digital,” “X-Work Digital Agency.” Consequently, every press placement, every Google Business Profile update, and every product description should use exactly the same brand name, category descriptor, and core positioning. This consistency is what allows AI systems to build a coherent entity signal for your brand.

4. Earn placement on comparison and roundup pages

A Trustpilot and Seer Interactive study found that brands that claimed a Trustpilot profile increased their AI citation rates from 1% to 54%. That finding is striking because it is not about quality of reviews — it is about presence on a category of sites that LLMs consistently cite. Review platforms, “best of” lists, and comparison articles represent 8.9% of all AI citations in Wellows’ 11.1-million citation analysis. For digital PR Malaysia strategies, this means actively pitching roundup features and ensuring the brand appears on relevant comparison platforms — not just mainstream media.

5. Target the publications LLMs already cite

The most efficient approach is to identify which Malaysian and Singaporean publications AI systems already use as citation sources for your category, then prioritise those for media relations. Run twenty to thirty buyer-intent queries across ChatGPT and Perplexity. Note which domains appear as cited sources. Then build a PR target list from those domains, starting with the ones that appear most frequently. This is research that takes less than a morning, and it replaces months of guesswork about media prioritisation.

How to Measure Whether Your PR Is Building AI Visibility

Traditional PR measurement — media impressions, advertising equivalency value — does not capture AI training signal. Therefore, measuring digital PR Malaysia results in 2026 requires a different dashboard.

Three metrics matter most:

Citation lift: Run a set of twenty to thirty non-branded queries relevant to your category across ChatGPT, Perplexity, and Google AI Overviews. Record which sources appear. Track whether your brand begins appearing as a cited source over time, and how often. This is not a one-shot test — AI citation visibility varies significantly between runs, so track across at least four to six weeks of consistent measurement before drawing conclusions.
AI share of voice: The percentage of relevant AI answers in which your brand is named, compared to competitors. A brand appearing in 12 out of 30 queries has 40% share of voice for that query set. Track this monthly and compare to three or four direct competitors.
Editorial citation rate: Of your media placements in the past quarter, how many were structured with named statistics, direct quotations, or inline citations? This is a leading indicator — placements built this way are more likely to appear in AI training data than those with vague, unattributed claims.

For brands doing SEO and AI search optimisation simultaneously, a clean measurement framework separates what drives traditional rankings from what builds AI citation authority. Both matter. However, they require different inputs, and conflating them produces a strategy that is diluted in both directions.

How This Connects to Malaysia’s GEO Landscape

The approach described here is sometimes called Generative Engine Optimisation (GEO) — the practice of structuring content and building off-page signals so AI systems are more likely to cite your brand. In the Malaysian market, most agencies are charging RM 2,000–3,000 per month extra for GEO as a standalone add-on on top of an existing SEO retainer. That model is questionable because GEO and SEO are not separable: the same editorial placements that build search authority also build AI citation signal. They run on the same fuel.

The distinction matters because it affects where you invest. A digital PR campaign in the AI citations landscape is not a luxury add-on. It is the activity that creates the independent, third-party evidence layer that both traditional search and AI search rely on to trust your brand. For Malaysian SMEs already investing in AI search visibility, digital PR is the off-page work that makes all the on-page work pay off.

As a practical starting point: identify two Tier 1 outlets from the list above that regularly cover your industry. Develop one piece of original research — a survey, a dataset, a benchmark — that gives journalists a citable number. Pitch it with a specific MY/SG angle. Then track your AI share of voice over the next sixty days. That sequence costs less than a month’s add-on GEO retainer at most Malaysian agencies, and it produces a result you can actually measure.

Frequently Asked Questions About Digital PR and AI Visibility

Does digital PR Malaysia actually influence what ChatGPT or Gemini says about my brand?

Yes, indirectly and over time. LLMs are trained on large web corpora that include news sites, trade publications, and community platforms. Brands that appear consistently in high-authority sources become part of the model’s understanding of their category. Real-time AI search tools like Perplexity and ChatGPT Search index new content within days of publication, meaning a fresh placement can influence answers faster than traditional SEO ranking changes.

Which media placements have the highest impact on AI training signal in Malaysia?

Placements in The Star, The Business Times, The Edge, TechInAsia, and Marketing-Interactive carry the highest AI training signal for the MY/SG market. The key factors are domain authority, crawling frequency, and whether the publication is already appearing as a citation source in AI responses to category queries. Run test queries on ChatGPT and Perplexity to identify which outlets your industry relies on before building a media list.

How is digital PR different from traditional link building for AI visibility?

Traditional link building targets Google’s PageRank signals — it is about backlinks and domain authority for web search. Digital PR for AI visibility targets the independent corroboration layer that LLMs use to decide what to cite. Whereas a link on a low-authority directory can move a traditional ranking metric, only editorially independent mentions in recognised publications build the kind of third-party validation that AI systems treat as trustworthy. The tactics overlap but the goals are distinct.

How long does it take for press coverage to improve AI citation rates?

Real-time AI search tools like Perplexity can surface new content within days of publication. LLMs that rely on training data updates — typically quarterly or semi-annual model refreshes — take longer. A reasonable planning horizon for meaningful citation rate improvement is three to four months of consistent media placements, combined with on-page content that meets AI citability standards.

Can Malaysian SMEs afford a digital PR Malaysia strategy, or is it only for large brands?

Digital PR scales down to SME budgets. The minimum effective unit is one strong piece of original research and two to three high-quality media placements per quarter. A survey of thirty to fifty Malaysian clients or customers can produce a statistically interesting finding. The placement cost depends on whether you are pitching independently or using a retained PR firm, but the content investment is within reach of any business already producing blog content. The additional step is making that content newsworthy and pitching it to the right outlets.

Does Xwork offer digital PR as part of its SEO services?

Xwork bundles GEO and AI search visibility work inside its Rank General protocol — not as a separate add-on. Off-page citation building, content structured for AI citability, and measurement of AI share of voice are part of the same engagement. If you want to understand how your brand currently appears in AI search results and what a structured programme to improve it would look like, an AI Search Visibility Audit is the starting point.