# AI bot - Agents

**URL:** https://meta.discourse.org/t/ai-bot-agents/306099
**Category:** Site Management
**Tags:** how-to, ai, ai-bot
**Created:** [April 30, 2024, 6:26am UTC](https://meta.discourse.org/t/ai-bot-agents/306099 "2024-04-30T06:26:18Z")
**Posts on this page:** 1
**Showing post:** 28

<div class="post-metadata">

### Author: ![philh](https://sea3.discourse-cdn.com/meta/user_avatar/meta.discourse.org/philh/32/532740_2.png) [@philh](https://meta.discourse.org/u/philh)
#### Post date: [February 14, 2026, 10:05pm UTC](https://meta.discourse.org/t/ai-bot-agents/306099/28 "2026-02-14T22:05:22Z")

</div>

Moving away from [Google Perspective API](https://meta.discourse.org/t/discourse-google-perspective-api/98733) and still getting my head around AI and the Discourse implementation of same.

What is the meaning/use for Allowed groups (optional) in the persona config? It seems to default to TL0  
 ![Screenshot 2026-02-14 140224](https://global.discourse-cdn.com/meta/original/4X/7/6/c/76c3b461ee0970be043004914a349d9c0e815bdb.png)

**Edit:** Perhaps elaborating will be useful.  
If Allowed Groups controls access or use then why is the default Spam persona set to TL0?

When would TL0 need access to the Spam Detector bot particularly when `AI spam detection max post count` and `AI spam detection max trust level` seem to be controlling control when the Spam Detector is applied?

Setting the Allow Group does make sense for say a group that is doing research and would need to access the Forum Researcher bot.

As always, what am I missing?

---

_[View the full topic](https://meta.discourse.org/t/ai-bot-agents/306099)._