Can You Use AI Translation for EMA eCTD Submissions
Yes, you can use AI translation for EMA eCTD submissions, on one condition: a certified regulatory linguist signs off on the output under ISO 18587, and the dossier never leaves EU-hosted infrastructure. Raw machine output pasted into a Summary of Product Characteristics and filed as-is is a different thing, and a national reviewer will treat it as one.
AD VERBUM is an EU-hosted AI translation provider that runs constrained large language model generation, then certified regulatory review, for eCTD Module 1 and EMA product information. We generate output on a client-tuned model held to your Translation Memory and Term Base, and a qualified linguist takes responsibility for the final text. Nothing trains on your data, and nothing leaves EU infrastructure.
The eCTD format itself does not care how a sentence was produced. What gets a submission sent back is content a reviewer reads against the approved English and the QRD template: a dosing line, a contraindication, a term rendered two ways across languages. Get the workflow right and AI translation is faster and steadier than the manual process it replaces. Get it wrong and you have filed an inconsistency finding with your name on it.
What EMA submissions actually require
The eCTD is the structured electronic format for EU marketing-authorisation dossiers. Modules 2 to 5 are the harmonised Common Technical Document, and Module 1 is the EU regional part, which holds the product information a patient and prescriber read. Since January 2010 the eCTD has been the only accepted electronic format for centralised-procedure submissions, so the language question sits inside a fixed technical envelope.
The product information moves as three linked annexes: the SmPC (Annex I), the labelling (Annex IIIA) and the package leaflet, or PIL (Annex IIIB). Directive 2001/83/EC Article 63 requires each national version in the official language or languages of the member state where the medicine is sold. For a centrally authorised product that means all 24 official EU languages, plus Icelandic and Norwegian for the EEA.
Timing is the pressure point. Within 5 calendar days of a positive CHMP opinion the marketing authorisation holder submits the final English product information plus every translation, and each member state checks its own version during the EMA linguistic review. No regulation names a translation tool. What EMA reviews is accuracy against the approved English and the QRD template, version by version.
What AI translation means for an eCTD dossier
AI translation in a regulated submission is not a synonym for machine translation. We run a large language model that reads the whole document, constrained by your approved terminology, and then a certified linguist post-edits and takes responsibility for the final text under ISO 18587, the standard for full human post-editing of machine and AI output.
Segment-level machine translation handles one sentence at a time, blind to the rest. That is how a negation flips inside a contraindication, and how posology drifts between two renderings across languages. A document-aware model with a governed Term Base makes far fewer of those errors. Fewer is not none, which is why the human sign-off is not optional. This AI+HUMAN hybrid translation model is what separates a submission-grade annex from a draft.
Where AI translation fits an EMA submission
Not every part of an eCTD dossier carries the same risk. These take AI translation well when a certified regulatory linguist reviews the output:
SmPC, labelling and package leaflet, where a governed Term Base holds excipients, posology and warning phrases steady across all 24 EU languages a submission reopens.
Module 1 administrative documents, cover letters and application forms, where standardised wording suits a constrained model.
Variations and periodic safety update reports, where translation memory reuses the unchanged bulk and only the delta is retranslated and reviewed, the same discipline we bring to rare-language pharma work.
Responses to questions during the procedure, where speed matters but the clinical wording still needs the certified review we apply to device IFUs.
The dividing line is content a patient or prescriber acts on directly. A dosing table, a contraindication, a safety warning: those get certified review every time, no exceptions.
What breaks eCTD compliance
Most of the risk people attribute to AI is really the absence of a few controls. These are what turn AI translation into a rejected submission:
Unreviewed output filed as-is, with no certified linguist accountable for the final text under ISO 18587.
Public-cloud transit of a dossier that still carries pre-approval safety data, instead of EU-hosted processing under the certifications regulated translation demands, held to ISO 27001.
Terminology drift across variations, where the same clinical term is rendered two ways because no shared term base held the approved equivalent.
A missing audit trail, so you cannot show a national reviewer which qualified linguist signed off each translated segment.
Missing the day-5 window in one language, which holds the entire submission or variation.
Miss any one and the workflow stops being defensible. The eCTD structure will accept the file. The linguistic review will not.
We built this for life sciences. Our certified regulatory linguists work under ISO 17100 and ISO 18587, on client-tuned open-weight models we host in the EU, and every SmPC, PIL and Module 1 annex moves through the same governed workflow we use for pharma clinical trials with the audit trail intact. For the wider picture, see our guide to AI translation companies for life sciences.
Our pharmaceutical translation services
Our pharmaceutical translation services run on ISO 27001 and ISO 42001 certified, EU-hosted infrastructure, with no reliance on public cloud tooling for core processing. Every project runs through our AI+HUMAN hybrid workflow: we ingest client Translation Memories and Term Bases first, our proprietary LLM-based LangOps System generates output constrained by client terminology on client-tuned open-weight models, and our certified subject-matter experts review for technical accuracy and regulatory compliance. Our QA is aligned to ISO 17100 and ISO 18587, with sector-specific requirements such as the QRD product-information template and Directive 2001/83/EC Article 63 language rules applied where relevant. We serve Life Sciences, Legal, Finance, Defense, and Manufacturing clients across 150+ languages with 3,500+ subject-matter linguists. For teams managing audit-sensitive content, contact us to discuss your security and compliance requirements directly.
FAQ
Can you use AI translation for EMA eCTD submissions?
Yes. The eCTD names no translation tool. It requires the product information in each member state's official language through a documented process, checked at linguistic review against the approved English and the QRD template. AI translation with certified post-editing under ISO 18587 meets that. Raw, unreviewed machine output does not.
Which eCTD documents can AI translation handle?
SmPC, labelling, package leaflet, Module 1 administrative documents, variations and PSURs, all with certified regulatory review. Anything a patient or prescriber acts on directly, like a dosing table or contraindication, gets full certified review regardless of how the first draft is produced.
Will my dossier be exposed to a public AI cloud?
Not with an EU-hosted provider. We run client-tuned open-weight models on EU infrastructure with no public-cloud API in the processing path, no training on your data, and no retention beyond contract, under ISO 27001. That matters while the dossier still carries pre-approval safety data.
How does AI translation keep terminology consistent across 24 EU languages?
A governed Translation Memory and Term Base lock approved terms before the model runs, so excipients, posology and warning phrases carry one signed-off equivalent per language. That is what stops a variation from reading as a divergence at linguistic review.
Does AI translation meet the EMA day-5 linguistic review deadline?
It helps. Within 5 calendar days of a positive CHMP opinion the marketing authorisation holder submits all language versions. Document-aware AI translation plus translation-memory reuse compresses the drafting time, while certified review under ISO 17100 keeps each version submission-grade.
What evidence does a national reviewer expect?
A documented audit trail linking each translated segment to a qualified reviewer, plus the ISO 17100 and ISO 18587 process records. The certification comes from that human sign-off against the QRD template, not from the model.
