Own the AI register and every AI file across the plants
- Where each model was tested, with its scores
- Independence proven by hash, not by assurance
- A sign-off queue with clocks per use
AI validation in pharma, designed around the draft EU GMP Annex 22: each model's test set laid over real production cell by cell and its independence from training proven by file hash. Criteria are signed before the test, subgroups are judged one by one, and a signed monitor pulls a drifting input back to manual work.
Watch one lab model tested, narrowed and signed ↓Data science builds a model for the lab, reports one headline accuracy, and hands over a test set drawn from whatever injections were already double-checked. Nobody lays that set over what the lab actually runs, so the newest instrument may have no test injections at all and the hardest method only a handful. Nobody asks the builder for the training manifest in a form that can be compared, so the test may reuse sequences the model learned from.
A classic CSV package of requirements, scripts and a trace matrix holds none of this. Criteria get written after the results are in, and one overall figure hides the method that fails. After go-live nothing watches the inputs: when a supplier changes a label layout, the reader starts guessing and an inspector notices first.
Built forCSV and IT compliance leadsQC lab headsQA approversMethod expertsHeads of quality
A headline accuracy hides the inputs a model was never tested on, and a test drawn from its training.
Evidence pieced together by hand
Built-in controls
A model goes live only on the inputs where it proved itself, and its signed monitor pulls it back when they drift.
An integration-check model meant to direct second-person review of chromatography: the CSV lead runs the re-check, the lab head limits the use, the method expert confirms it and QA signs.
demoSample data: the plants, clients, people, models, injections and results here are invented.
Who buys and runs the work in a lab or plant that puts models into GMP decisions.
Own the AI register and every AI file across the plants
Act as system owner: sign what the model may and may not decide
Sign each AI file, each model change and each learned rule
Set acceptance criteria before the test and confirm the described use is adequate
Around the file: the register and its queue, blind labels, requests to builders, change control, a generative assistant and the trail.
demoSample data, from the working solution.
Seven agents map, check, run, challenge, compile and watch; people sign the file.
Drafts what the model decides, never decides and who acts on its answer, from the SOP, the data system process and the builder's notes.
Counts production by method and instrument against the test set, flags thin and empty cells, and tops them up under the test plan.
Asks the builder for the manifest as fields, compares sequence IDs, injection IDs and file hashes, and records every exclusion.
Runs the frozen model after checking its hash, counts results per subgroup, and logs confidence and feature maps.
Asks an inspector's questions of the test set and of the file the compiler drafted.
Builds the file in the draft annex's section order, each section citing its records, once the method expert confirms the use.
Runs each signed monitor plan weekly and returns a drifting input to manual work when its trigger fires.
AI validation in pharma is the evidence that a model used in GMP work does what its intended use says, on the inputs it will really see. That means a test set kept apart from training, labels checked by two experts, criteria set before the test and results per subgroup; here agents build the evidence and people sign it.
It went to public consultation beside a revised Annex 11. For static, deterministic models in critical GMP uses it asks for an intended use with its input sample space, subgroups, criteria approved before testing, independent and verified test data, confidence thresholds, change control and drift monitoring. Generative models stay out of critical uses.
The model's builder returns the training manifest as fields: sequence IDs, injection IDs, file hashes and the freeze date. The checker compares them with the test set, replaces every overlap with injections acquired after the freeze, and records each exclusion with its reason. Builders get no access to the test data.
The file says so on page one. The system owner rehearses the options and can narrow the intended use, keeping that method on full human review, or send the model back to its builder for a new version and a new test. QA signs only within the stated use.
Each live use runs under a signed monitor plan with its own metric and trigger, such as the undecided rate per material for a label reader. When a trigger fires, that input goes back to manual work under the plan, and a model update needs change control and a fresh test.
They sit on the same register, designed as non-critical: a trained person writes, edits and owns every output, the assistant has its own test set, and a monitor tracks how often people change its drafts.
Security fixes go in on their clock, and each function a patch reaches is tested with a reason an auditor can follow.
1-min demoComputer Software AssuranceTesting sized to each function's risk, every line linked to its source.
1-min demoAudit Trail ReviewEvery audit-trail period reviewed on schedule and signed, rows cited
1-min demoSend its intended use, its test set and the training manifest; we map the test against production and check independence by hash.