> For the complete documentation index, see [llms.txt](https://opora.gitbook.io/opora-health-documentation/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://opora.gitbook.io/opora-health-documentation/opora-core-for-health-ai-funding/validation-and-technical-robustness-of-the-ai-enabled-system.md).

# Validation and technical robustness of the AI-enabled system

If your system makes or supports health-related decisions, funders expect you to show **how you check that it works as intended, and how you know when it stops doing so**. It is about demonstrating that your system can be trusted during development and use.

#### What you need to show

* Which data was used to train, test, and validate the system.
* How performance is measured and tracked over time.
* How model changes are controlled and documented.
* Where humans intervene and how those decisions are recorded.

#### How OPORA Core supports this

OPORA Core provides:

* Defined validation workflows matched to your current maturity level.
* Clear separation and tracking of training, validation, and test data.
* Versioning of models, datasets, and results.
* Logging of human review and override actions.
* System-level monitoring for performance changes and data drift.

This allows you to point to real system evidence.