The Segment Built for Breaking News Is Not Where the Press Wins
Six categories in the Machine Relations Index publish a News-driven citations segment, built specifically for breaking-news queries. Across their top-ten slots, vendor-owned domains hold 22, editorial publications hold 13, and wire and press-release distribution holds zero.
Six categories in the Machine Relations Index publish a segment built for exactly one kind of query: news-driven citations, the shape reserved for breaking developments inside an industry rather than steady-state buyer research. It is the segment where a press release, if wire distribution earns citations anywhere, should earn them. Across the sixty top-ten slots those six segments hold between them, vendor-owned domains occupy twenty-two, editorial publications thirteen, and the wire and press-release distribution class — the class built to feed exactly this kind of query — zero.
Paloaltonetworks.com ranks #1 of 1,340 cited domains in Cybersecurity's news-driven segment, cited in 81 of 623 observed runs. PR Newswire is cited in the same segment, ranked #46, and it is the highest-ranking wire domain in any of the six.
What was measured
Every figure here comes from the public release at machinerelations.ai/data/machine-relations-index.json, contract machine_relations_index_public_view_v2.0, methodology mri_score_v2.0, artifact 47973f373a20, generated September 20, 2026. The observation window runs May 10 to September 20, 2026 — 127 days, 16,039 observed answer runs, 919 monitored prompts, 126,271 citation events, 22,320 cited source domains, six engines.
The six are ChatGPT, Claude, Gemini, Google AI Mode, Google AI Overviews and Perplexity. Each publishes its own account of how it surfaces and links sources: OpenAI documents a search crawler distinct from its training crawler (OpenAI bots), Google describes AI Mode as its own search experience (Google, AI Mode) and documents its crawler fleet and AI-surface guidance separately (common crawlers, AI features and your site), Anthropic documents citations as part of Claude's web search (Anthropic, web search tool), Google documents grounding with Search for Gemini (Gemini API), and Perplexity documents its own retrieval surface (Perplexity). Each reaches the web under its own user agent, governed by the Robots Exclusion Protocol as standardised in RFC 9309.
A segment is one subject category paired with one buyer question shape. This release carries 157 strata, of which 151 are collectable, 90 are published and 61 are still collecting.
A seventh question shape, built for a different question
Every category in the Index is measured on six standing buyer-question shapes: best tools, how buyers choose, is it worth it, problem-first research, top lists, comparisons. A seventh shape exists in only some categories: news-driven citations, sized differently from the other six by design. Where a standard segment clears its evidence floor at 7 distinct run dates, the news-driven segments in this release carry 53 or 54 distinct dates — roughly the whole window — because the shape is built to sample a steady stream of "what just happened" queries rather than one static buyer question.
Six categories have cleared the floor on this shape in the current release, each already carrying hundreds of runs and over a thousand citation observations:
| Category | Observed runs | Run dates | Cited domain observations |
|---|---|---|---|
| Cybersecurity | 623 | 54 | 1,340 |
| Enterprise Software | 596 | 53 | 1,375 |
| Fintech | 616 | 53 | 1,531 |
| Healthcare Services | 613 | 53 | 1,264 |
| HR & Talent | 608 | 53 | 1,249 |
| Martech & Advertising | 602 | 53 | 1,244 |
A seventh category, Legacy News Topics, also publishes this shape and is the largest single segment in the release at 2,138 runs across 54 dates — it is not included in the leaderboard analysis below because its per-domain segment page was not read for this piece, and every figure here is scoped to the six categories actually read.
This is the shape a wire story is built for: a company announcement, a breach disclosure, a funding round, a regulatory filing, something that just happened and that a newsroom or an AI answer engine is choosing a source for right now. If wire and press-release distribution earns citations anywhere in the Index, this segment is where the case should be strongest. PR Newswire's own description of its service is distribution to media outlets, regulatory systems and search and AI engines — a claim this segment is a direct test of.
Where the top-ten slots actually sit
Taking the top ten ranked domains in each of the six categories' news-driven segment — sixty slots in total — and reading off the source role the release assigns to each:
| Source role | Top-ten slots held | Share of 60 |
|---|---|---|
| Vendor-owned source | 22 | 36.7% |
| Editorial publication | 13 | 21.7% |
| Community and social platform | 9 | 15.0% |
| Analyst and consulting research | 8 | 13.3% |
| Market and company database | 5 | 8.3% |
| Academic and government source | 3 | 5.0% |
| Wire and press-release distribution | 0 | 0.0% |
Vendor-owned domains hold more top-ten slots than editorial publications by a wide margin — 22 against 13, from a class that is a third the size of editorial across the Index as a whole (823 vendor-owned domains against 1,230 editorial publications in this release). Wire and press-release distribution, a nine-domain class across the whole Index, holds none of the sixty.
The rank-1 position in each category goes to a different kind of source, which is itself informative — this is not one class running the table everywhere, it is vendor-owned sources accumulating the largest share across the field. Ranks below follow standard competition ranking, where tied domains share the better rank rather than being re-sorted apart:
| Category | Rank-1 domain | Source role | Citation rate | Cited runs |
|---|---|---|---|---|
| Cybersecurity | paloaltonetworks.com | Vendor-owned source | 13.00% | 81 of 623 |
| Enterprise Software | linkedin.com | Community and social platform | 11.07% | 66 of 596 |
| Fintech | linkedin.com | Community and social platform | 10.23% | 63 of 616 |
| Healthcare Services | nih.gov | Academic and government source | 25.12% | 154 of 613 |
| HR & Talent | unleash.ai | Editorial publication | 11.35% | 69 of 608 |
| Martech & Advertising | martech.org | Editorial publication | 15.12% | 91 of 602 |
No wire domain leads any of the six. Cybersecurity is the clearest single-class win: the top eight slots include five vendor-owned domains — paloaltonetworks.com, sentinelone.com, microsoft.com, crowdstrike.com and fortinet.com — before the first editorial publication, securityweek.com, appears at rank 7.
Where the wire class actually sits
Wire and press-release distribution domains are cited in five of the six segments, just never inside the top ten:
| Category | Highest-ranking wire domain | Standing | Citation rate | Cited runs |
|---|---|---|---|---|
| Cybersecurity | businesswire.com | #58 of 1,340 | 2.25% | 14 of 623 |
| Enterprise Software | none observed | — | — | — |
| Fintech | financialcontent.com | #41 of 1,531 | 2.76% | 17 of 616 |
| Healthcare Services | businesswire.com | #72 of 1,264 | 1.79% | 11 of 613 |
| HR & Talent | globenewswire.com | #55 of 1,249 | 2.47% | 15 of 608 |
| Martech & Advertising | none observed | — | — | — |
Cybersecurity's segment also cites PR Newswire at #46 and GlobeNewswire at #95; the Index classifies PR Newswire itself as an editorial publication rather than wire and press-release distribution, a distinct question from whether its citations concentrate in this segment. In Enterprise Software and Martech & Advertising, no domain the release assigns to the wire and press-release distribution class appears anywhere on the leaderboard at all — not absent from the top ten, absent from the full ranked field of 1,375 and 1,244 domains respectively.
This is not a contradiction of the segment's own premise. News-driven citations still cite sources that report on an event — securityweek.com, infosecurity-magazine.com, hrexecutive.com, fintechfutures.com, medcitynews.com are all present in these leaderboards. What the data does not support is the assumption that a wire release itself, as the distribution mechanism, is what an engine cites when the underlying query is exactly the kind of breaking-news question a wire release is filed for. The vendor whose product the story is about, the analyst firm that covers the category, and the trade publication that reports on it all rank ahead of the distribution channel in this release, in every one of the five categories where a wire domain appears at all.
What this does not mean
This is one segment, in one release, over one 127-day window. It does not measure whether a specific company's wire release was picked up, quoted, or linked from a cited page — the Index measures which root domains are cited, not the provenance chain behind any single citation. It also does not measure earned media placement, which is a distinct question from the domain an engine ultimately cites when answering a news-shaped query.
It does not establish that wire distribution has no editorial value; a release can still be the origin of a fact that surfaces on a cited domain further down the chain, and that chain is not visible in this dataset. What it does establish, on the domains actually cited, is that in the specific segment built to test this exact case, the wire and press-release distribution class holds no top-ten position in any of six categories, and no position at all in two of them.
The population is cited domains, not eligible ones. A domain cited zero times in the window is not in the file. And a citation is not proof that the cited page supported the claim beside it — that is measured separately in Answer-Source Fidelity, and none of the results above follow from it.
Release: mri_score_v2.0, contract machine_relations_index_public_view_v2.0, artifact 47973f373a20, generated 2026-09-20. Window: 2026-05-10 to 2026-09-20, 127 days observed. Segments read: the News-driven citations leaderboard for Cybersecurity (623 runs, 54 dates), Enterprise Software (596, 53), Fintech (616, 53), Healthcare Services (613, 53), HR & Talent (608, 53) and Martech & Advertising (602, 53), each read from its own published segment page. Evidence floor: 10 observed runs across 7 distinct run dates, applied to a stratum. See also the wire and press-release concentration finding in the whole-Index distribution curve.