Hi, I have configured the “Triage posts using AI” automation for content moderation on my platform (madeira.community — a multilingual business directory for Madeira island). The automation works perfectly for public posts. However, I enabled “Include personal messages” option and found that DMs between regular users are not being scanned at all. My concern is real: bad actors can use private messages to conduct illegal activities — drug dealing, sharing stolen financial data, grooming — while completely bypassing AI moderation that works fine on public content. My questions: Does “Include personal messages” actually support scanning DMs between two regular users (non-staff)? If not — is this by design or a known limitation? Is there any supported way to apply AI triage to private messages between regular users? I found a reference to a commit “Skip PM scanning in LLM triage by default” which suggests this may be intentional. But the UI option exists — so I am confused about what it actually does. Thank you

I can confirm that they were not processed, the topic ID’s of those PM’s (210, 211, 212 and 214) and post id’s are not in the ai_api_audit_logs table. I think “Topic Type” “All topics” means all topics that are not private messages. Could you try to leave that one empty?

[图片] Denis_Kovalenko: Does “Include personal messages” actually support scanning DMs between two regular users (non-staff)? Yes it does, if you enable that option it should be scanning PM’s. [图片] Denis_Kovalenko: and found that DMs between regular users are not being scanned at all. How did you determine this?

“I tested it by sending a private message between two regular users (testUser to darinauser, both non-staff) with clearly illegal content: ‘кокаин. оружие. гранаты’ and ‘порно. грязь. разврат. насилие’. The messages were delivered and not flagged. The Review Queue remained empty. The same content in public posts was flagged immediately.” [1.1] [PM1.2] [PM1.1] [promt] [2.1]

"That worked! Thank you Richard. Leaving the Topic Type empty fixed it — DMs between regular users are now being scanned. I’m now planning to set up two separate automations: One for public posts with a strict moderation prompt One for private messages with a lighter prompt — flagging only the most serious violations (CSAM, drug dealing, weapons) This way public content gets full moderation while private conversations have more appropriate boundaries." the topic can be closed. THX!

Hmm “all topics” does seem a little unclear in this context… maybe we should update to clarify all topics topics (all categories) Public Topics (exclude all secure categories) topics (public categories only) personal messages only

The “All Topics” part made me think all content would be reviewed. I didn’t even consider that leaving the field blank would result in all content being reviewed.

I think having a single dropdown is pretty limited and three separate options would be better and definitely more clear to a user. public topics yes/no non-public topics yes/no private messages yes/no There are 8 possible situations and the drop down only covers four of them.

maybe so ? All ( topics + pm ) public topics only non-public topics only non-public + pm private messages only

That does not cover “regular topics ( both public and non-public), but no PM’s” which would be a pretty logical choice. And, albeit a less likely choice, “public topics, and PM’s”

AI 分诊自动化是否会扫描普通用户之间的私信？

产品支持

Denis_Kovalenko (Don't worry - Be happy) 2026 年4 月 6 日 16:47 9

也许吧？

全部（主题 + 私信）
仅公开主题
仅非公开主题
非公开主题 + 私信
仅私信

话题		回复	浏览量
Discourse AI - Spam detection Site Management moderation , how-to , ai , spam	31	6238	2026 年8 月 28 日
Experiments with AI based moderation on Discourse Meta Community Building moderation , ai	11	1834	2025 年5 月 26 日
Listing conversations with artificial intelligence on a separate page or filtering them on the messages page Feature ai	9	494	2026 年1 月 10 日
Allow users publish AI Bot PM as new topic Feature completed , ai , ai-bot	18	1520	2023 年12 月 30 日
Thoughts on Turning Off Personal Messaging and Chats Community Building chat , personal-messages	35	1929	2026 年1 月 3 日

AI 分诊自动化是否会扫描普通用户之间的私信？

相关话题