I’m seeing more and more spam on the site.
- Is there any way we can handle this better?
- Is there any way to bulk flag it?
2 Likes
On one occasion, discourse said I couldn’t flag any more - wait 12h before flagging spam. If no automated mechanism can be implemented, at least raise the limit on the available spam arrival intensity countermeasure, flagging.
I’ve been using Gemini as a proactive content moderator for my Discord communities (Something I specialize in)
Is there possibly any developer APIs for Discourse including Webhooks? It’d be great to see Google Gemini protecting its own forum.
If a staff member or moderator is viewing this, possibly try something like this?
Now that I’ve actually looked into the issue - It’s bad. Found this MASSIVE list of spam bots, and they’ve all got their birthday as January First.
Maybe start looking for patterns like these? I’m really saddened that about half of the accounts on here are bots.
2 Likes
I don’t know how - but these spam “WhatsApp” posts are being re-listed???
I am clueless on how/why these are being relisted, even though they are clearly spam.
The worst thing is - is that when these get de-listed, they aren’t even de-listed?? They still show up under Latest as the spam-bots reply to their own post with the same junk. If something is de-listed, it should de-list the whole thread (as long as the main post is de-listed).
Google, surely it wouldn’t be that hard to implement a fix for this. Your own models can solve the issue for you. It even found two that I’ve managed to flag and get de-listed with a really simple prompt
For those that do have Discourse themselves, it’s really not that hard. Discourse gives you a really easy way to implement this - their public API shows a way to get posts and remove them really easily.
Please, Google. Help us fix this problem once and for all.
1 Like