# AI & data breaches?

**URL:** <https://community.fibery.io/t/ai-data-breaches/10550>\
**Category:** Ideas & Features\
**Created:** [March 13, 2026, 1:27pm UTC](https://community.fibery.io/t/ai-data-breaches/10550 "2026-03-13T13:27:34Z")\
**Posts on this page:** 1\
**Showing post:** 6

<div class="post-metadata">

**Author:** ![papertroll](https://sea2.discourse-cdn.com/flex020/user_avatar/community.fibery.io/papertroll/32/14157_2.png) [@papertroll](https://community.fibery.io/u/papertroll)\
**Post date:** [July 24, 2026, 2:55am UTC](https://community.fibery.io/t/ai-data-breaches/10550/6 "2026-07-24T02:55:22Z")

</div>

Just adding a +1 on this topic.

Given that AI Search is much more powerful than standard Fibery search, it is unfortunate that there is no way to run it through a self-hosted (by user or by Fibery) model as @myg_ge suggested, or to ensure zero data retention. Currently, I cannot use Fibery’s AI search due to org policies. It would be great if there were “_solutions […] so that AI doesn’t “steal” the sensitive data?_” as @myg_ge said, that aren’t just to disable AI.

**To clarify my request:** it would be very useful to have either a) ability to use a self-hosted model for semantic search or, as a less ideal but probably easier-to-implement solution, b) Fibery self-hosts the model for semantic search with zero data retention OR gets a zero data retention agreement with the subprocessors.

I would assume many other Fibery users (or potential Fibery users) would like a solution to this too, but maybe I’m being selfish 😅

Thanks for considering!

* * *

**Other relevant comments, for reference**

For clarify, I can see this has been discussed before here:

> [@Fibery Product Report: September 2024](https://community.fibery.io/t/fibery-product-report-september-2024/7572/1):
>
> Grand idea is to host own model for embeddings and get rid of third-party services like OpenAI. In this case we can index databases by default and provide great search experience for all accounts. However, in the first steps we will try to improve search quality and usability, and only then think about self-hosting.

And here:

> [@Feature Request: local large model for AI Search](https://community.fibery.io/t/feature-request-local-large-model-for-ai-search/5953/3):
>
> Not in near future unfortunately. We want to do it at some point though

And in the [Expose Semantic Search through MCP](https://community.fibery.io/t/expose-semantic-search-through-mcp/10999) request

---

_[View the full topic](https://community.fibery.io/t/ai-data-breaches/10550)._