Everything we publish, in one place.
Guides, whitepapers, articles, glossary, FAQs, and product comparisons — the full library of how Asthra thinks about AI in regulated writing. Start with a guide to learn the craft, a whitepaper for the long argument, or the blog for the running commentary.
Long-form positioning documents — the argument for why regulatory writing needs a different stack, and what that stack looks like.
Where Inference Enters the Record: Governing AI in Regulatory Writing
A sophisticated regulator is not asking whether you used AI. The question is where inference entered your document, and who owns it. How Asthra builds for that question — five modes of AI use, a control that fails closed, and a prompt-review failsafe.
Beyond Frontier Models — Why Regulatory AI Needs a Purpose-Built Stack
Frontier LLMs and generic RAG aren't enough for regulatory writing. A purpose-built stack — document model, retrieval, planning, and verification — is what separates a demo from a draft a regulator will accept.
Why Regulatory Writing Is Broken — And How Agentic AI Fixes It
Regulatory writing teams spend 60–70% of project time on mechanical document assembly while sponsors lose $1.4M per day of delay. Why generic AI tools fall short, and what an agentic, document-aware system actually changes.
Evergreen, plain-English guides to regulatory writing — the documents, the ICH guidelines behind them, and how to get into the field. The reference we wish existed.
How to Write an FDA Meeting Package (Briefing Book)
The meeting package is the document that earns your FDA advice. Here is what belongs in it, why the questions are capped at ten, how the numbering rule works, and when each meeting type's package is due.
Read guide →Breaking into Regulatory Writing from the Lab, Pharmacy, or Clinic
Many regulatory writers came from the lab, pharmacy, or clinic. What transfers from each background, what's missing, and the concrete bridge move for each.
Read guide →CSR vs Clinical Overview: Two Very Different Documents
A CSR reports one trial in full under ICH E3 (Module 5); the Clinical Overview is a short benefit-risk argument for the whole program (Module 2.5).
Read guide →Articles, pilot debriefs, and regulatory news — written for medical writers, regulatory leads, and the people who hold them accountable.
Correct numbers, wrong population
Our drafting agent used subgroup results to describe overall safety. Our automated evaluator missed it. What this taught us about checking the claim behind the numbers.
Read article →A Bayesian Protocol Has to Argue for Its Prior. FDA's Draft Guidance Sets Out What That Argument Contains.
FDA's January 2026 draft guidance on Bayesian methodology carries a documentation section that asks the protocol to justify the prior distribution, describe every external source that informed it, name the sources excluded and why, and account for evidence suggesting skepticism. The writing burden is provenance.
Read article →'Avoid Concomitant Use' Over 'Use With Caution': FDA Is Specifying the Prose in Section 7
FDA's draft guidance on drug interaction labeling tells applicants which phrasings to prefer, what order the sentences go in, when to use a table, and to write in the active voice. Two agency webinars on 23 September walk through it. It is one of the most explicit pieces of editorial direction FDA has published for a single label section.
Read article →Definitions, common questions, and side-by-side comparisons against other tools.
Events
Where to meet Asthra in person — including the DIA Whale Tank at DIA 2026 in Philadelphia.
See events →Glossary
The terms regulatory writers and AI engineers use when they talk past each other — defined.
Open glossary →FAQ
Quick answers to the questions sponsors and writers ask before they pilot.
Read FAQs →Compare
How Asthra compares to generic copilots, in-house templates, and other regulatory AI tools.
See comparison →See Asthra on your documents.
Request a demo and we'll match Asthra's draft against your existing process — section by section.