# Should testing data be made available to \*use\* an AI system?

**URL:** <https://discuss.opensource.org/t/should-testing-data-be-made-available-to-use-an-ai-system/139>\
**Category:** Open Source AI\
**Tags:** draft\
**Created:** [January 26, 2024, 7:42am UTC](https://discuss.opensource.org/t/should-testing-data-be-made-available-to-use-an-ai-system/139 "2024-01-26T07:42:16Z")\
**Posts on this page:** 1\
**Showing post:** 10

<div class="post-metadata">

**Author:** ![gvlx](https://yyz1.discourse-cdn.com/flex011/user_avatar/discuss.opensource.org/gvlx/32/381_2.png) [@gvlx](https://discuss.opensource.org/u/gvlx)\
**Post date:** [September 26, 2024, 9:04am UTC](https://discuss.opensource.org/t/should-testing-data-be-made-available-to-use-an-ai-system/139/10 "2024-09-26T09:04:28Z")

</div>

(Reopening this topic, as it is becoming increasing relevant to current discussions [Draft v.0.0.9 of the Open Source AI Definition is available for comments - #15 by Shamar](https://discuss.opensource.org/t/draft-v-0-0-9-of-the-open-source-ai-definition-is-available-for-comments/513/15) )

I beg to disagree.

If you want to validate the claims of a particular Ai system (as you would if you have the freedom to study it), you will absolutely need _all_ the original data it was used to test it.

It does not matter that you _can_ create a new test dataset which could invalidate the original claims, you _need_ to test the original conclusions against the original (train and test) data.

As with any other Open Source product, _most_ people are only users; _some_ will compile it; a _few_ will study it and _develop_ new features or correct bugs.

You need to have access to all parts to be able to exercise your four freedoms.

---

_[View the full topic](https://discuss.opensource.org/t/should-testing-data-be-made-available-to-use-an-ai-system/139)._