How we report

What a Regulatory News story is made of, who is accountable for it, and where the machines fit. Written plainly, because you are entitled to know how the thing you are reading was made.

Most publications tell you their standards. Fewer tell you their process. The difference matters more than it used to, because the tools available to a newsroom in 2026 are not the tools that were available in 2016, and a reader has no way of knowing which ones were used unless the publication says so.

So here is ours, in full.


1. Every story starts at the primary source

A story begins with a document: a Federal Register notice, an FDA database record, an EMA committee agenda or opinion, an MHRA safety alert, an Official Journal entry, a company’s own release or filing. Not another outlet’s write-up of it. Not a press summary of a summary.

If we cannot reach the underlying record, we do not run the story. Every factual claim in a published piece links to the document it came from, so you can check us without taking our word for anything.

2. Drafting is AI-assisted, and that is a deliberate choice

Regulatory News uses an internal publishing pipeline that reads those primary documents and produces a structured first draft against a fixed template — the summary, the facts table, the timeline, the comparison, the questions a reader is likely to have. The pipeline is ours; it is not a chatbot asked to write a news story, and it is not given a topic and left to invent the details.

We use it for a specific reason. Regulatory activity is high-volume, highly structured and unglamorous: dozens of clearances, opinions, designations and notices a week that professionals need to know about and that no outlet covers because covering them by hand does not pay. A pipeline that reads the record and lays out what it says lets a very small newsroom cover a very wide beat, at a standard of sourcing that is higher, not lower, than the industry norm — because the source link is generated with the claim rather than added afterwards if someone remembers.

What the pipeline does not do is decide what is true. That is the next step, and it is not optional.

3. A human reviews and publishes every story

Nothing publishes itself. Every story is reviewed before it goes out by Connor Griggs, the publication’s editor — a regulatory affairs professional who began his career as a regulatory analyst at the U.S. Food and Drug Administration and has spent thirteen years in medical device regulatory affairs and quality since. His byline is on every story because he is accountable for every story.

Review means checking the claims against the cited record, correcting what the draft got wrong or framed badly, cutting what cannot be supported, and adding the context that only comes from having done the work. A draft that does not survive that review does not run.

4. What we will not do

  • We do not publish a claim we cannot trace to a named, linkable primary source. No unattributed figures, no “reports suggest”.
  • We do not invent quotes. A quotation in a Regulatory News story is a quotation from a document or a release, and it links to it.
  • We do not silently edit a published story. Corrections are appended, dated, and listed on the corrections page.
  • We do not let a sponsor near the newsroom. Paid placements are labelled, are attributed to the sponsor rather than the editor, and are governed by the sponsored content policy.
  • We do not dress automation up as legwork. We have not conducted an interview we did not conduct, and where a story rests entirely on the documentary record — which is most of them — that is what it says it rests on.

5. What this means for you

Treat a Regulatory News story as a well-sourced, human-reviewed reading of a public document, not as independent field reporting. Where a decision matters to your programme, follow the link to the record and read it yourself. That is what the links are for, and it is the reason we put them on every claim rather than at the bottom.

Why say any of this? Because the alternative is letting you assume something untrue about how these stories are made, and a publication that is not straight with its readers about its own process has no standing to hold anyone else to theirs.


Questions, or something we got wrong

Email editor@regulatorynews.com. If it is a factual error, say which story, which sentence, and which primary source shows it is wrong — corrections get read before anything else in the inbox.