This post by @MeltedMottle criticizing the whole Google/GenAI thing was flagged by the Discourse AI and taken down. No reason was given. I don’t want to make any definitive accusations, but this whole thing reeks of censorship. Original post was not hateful or anything, the user was just very concerned about the consequences of iNat partnering with a notoriously unsustainable corporation. Even if you disagree with OP, there was no reason for their post to be taken down, especially by the “Discourse AI” of all things. There should be room on these forums to discuss, rather than shutting people down. If any iNat mods know anything more about this, please reply. I love this community and I don’t want to see it destroyed by censorship and division.
this is not correct
The post was not taken down by Discourse AI, it was flagged by multiple other non-staff/non-moderator forum members, and hence it was hidden. Nothing mysterious about it
The reference to Discourse’s spam filter in the last comment of the linked thread refers to that user being flagged/silenced by the Discourse filter for a different comment they made on another thread; Tony was using the neutral pronoun ‘it’ to refer to said user, which you interpreted as referring to the post itself.
According to Discourse staff, the AI spam filter works in the following way:
What content gets scanned?
AI spam detection evaluates posts based on these criteria:User trust level:
Scans posts from users with trust level 1 or lower.
Excludes posts from higher trust levels.
Post type:Public posts (excluding private messages).
Both reply posts and first topic posts are included, based on additional thresholds.
Post edits:Scans posts with significant edits (e.g., changes exceeding 10 characters).
Enforces a 10-minute delay between scans of the same post.
Post frequency:Prioritizes posts from new users with fewer than 4 total posts in public topics.
Excludes posts from users exceeding this threshold.The classification process
Posts that meet the criteria are sent to an AI model (LLM) for analysis. The model evaluates whether the post is “SPAM” or “NOT SPAM” based on:Context: Includes post content, topic title, user account data (e.g., account age and trust level), and site guidelines.
Custom instructions: Admin-defined rules for reinforced or adapted scanning criteria.
Automated detection:
Flags irrelevant or promotional content (e.g., ads or commercial materials).
Identifies automated or bot-like behaviors.
Assesses content relevance to the discussion.
My post got flagged and hidden too. I never thought there was much censorship at iNaturalist. Guess I was wrong.
How many?
enough to trigger Discourse’s software to automatically hide the post
Thank you for the clarification, however I still do not understand why the post was flagged by other users. The post was not spam and in fact was very thoughtful and well-written. I understand that the user’s opinion was divisive but that was no reason to delete the post, instead all the more reason to leave it up so that other users could discuss.
If you have a problem with getting many of your posts flagged or hidden. I honestly suggest analyzing most importantly how you are saying what you’re saying, and to a much lesser extent what.
I feel i have no issue with my posts getting hidden. I try to be respectful in how I say things and to follow the forum guidelines. This doesn’t mean I/you can’t be critical of issues with INaturalist. I have plenty of comments that are critical of the iNaturalist CV in particular.
Also a large chunk of the comments in that forum post are critical of the generative AI thing anyways.
there are multiple different flag reasons that users can raise on a post, spam is only one of them. The post in question was not flagged for spam
the post was not deleted, it was hidden
I apologize for using the incorrect terminology, but I believe my point still stands.
how can someone find out their “trust level”. Who determines the trust level—is it Discours? How can LLM determine what is spam? These actions make me less satisfied with the AI iNat-Google plan than I was before.
You can view your trust level on your forum profile. TL0 - New User, TL1 - Basic user, TL2 - Member, TL3 - Regular. This page shares the gist, but the iNat Forum doesn’t have all the same permissions per trust level that the page references: https://blog.discourse.org/2018/06/understanding-discourse-trust-levels/
You can learn more about Discourse spam detection here: https://meta.discourse.org/t/discourse-ai-spam-detection/343541
If you have other general questions related to Discourse (the software this forum uses) there is meta.discourse.org - or if you have questions related to the iNat forum’s implementation of it specifically you can create a new topic at Forum Feedback
For what it’s worth, the account in question is able to edit their post and/or reach out to moderators about any moderation actions but has not done so yet.
For more about Forum guidelines and various moderation tools used here, you can read the Forum Guidelines page, which is also linked to on the sidebar.
If you peruse many discussions here you’ll find there’s plenty of criticism of iNaturalist.
I flagged your post as being off-topic because it was a complaint about Google Maps and not about the Google.org grant/AI. You should be able to edit it if you like and make it visible again. But it may be flagged again if it’s still off topic. See the Forum Guidelines.
Hiding posts: to remove inappropriate content from view, such as personal attacks, bad faith arguments, trolling, off-topic posts, and repetitive posts coming from the same user. The poster should be able to edit the post after a 10 minute waiting period, and the newly-edited post will appear again. If the post is flagged after that, it will be hidden unless a moderator or staff member unhides it.