This is definitely true, but it’s not like there’s anything anyone here can do about it; it’s not really fair to make @anon11892322 have to answer for stuff anyone else said or did.
This is mis-characterizing @nipos previous posts on LLMs and their output pretty wildly, and seems to try to paint them as an apartheidist? It certainly attacks the person, rather than their words.
If I’m misunderstanding your intent, can you please try to be more careful? Personal attacks aren’t helpful and anything that looks like one is likely to be interpreted as having been meant that way.
1 Like
Maybe someone should fork Haiku and create an AI allowed version of it. Let’s call it Dano. Devs against AI can simply continue their work without any AI related discussions. The other devs can go their merry way and still keep up-to-date with Haiku, for the time being.
1 Like
41 posts were split to a new topic: On The Merits Of LLMs
Hi folks.
Y’all very clearly want to go off into the weeds on the merits or lack thereof of LLMs.
Fine. 
But that is off-topic for the forum, so I’ve split the stuff after this point into an Off-Topic thread.
Let’s try to keep this thread on topic; it’s about moderation style here and whether it’s too heavy handed …
1 Like
Neither. The Initial post was marked as spam, as it was yet again generated by an LLM, the forum then automatically hid the topic. There was no attempt to edit the initial post to remove the LLM spam, hence the topic stayed hidden…
But besides that, I don’t see the point. We don’t need an audit trail. If you want to know specifcs you can just ask, and you will be answered. I don’t think we should keep LLM spam up publically for search engines to index.
Not directly an answer to you, but this topics Original thesis is kind of ridicilous. There never was any “silencing”, users are free to open topics or send a pm to the moderation team… this Whole topic is evidence of that; but it was not opened as an “Can someone explain this?” or an answer to waddlesplash or anything, instead it was a direct attack on how dare waddlesplash add context to a closed topic .-. (beeing transparent about mod stuff is a bad thing apparently)
Indeed, it seems it was opened to Yet again talk about LLMs, the thinly veiled “concern” about moderation was dropped quite quickly.
EDIT: to answer your initiial querry, he was banned for a month for his initial topic where he attacked the Haiku community, getting with the genius yellow star analogy for marking LLM generated software; and harrassing another developer with LLM generated images
2 Likes
@VoloDroid The ban you remembered is the 30-day one, not the 24-hour suspension. And yes, the apology post you remember did exist: I posted it a few days ago in the thread that was later closed, and it was specifically an apology for the “yellow star” image. I don’t know whether it was removed by the moderators or auto-hidden by the forum software; I can only confirm it was there.
I’ll repeat that apology here, since the point keeps coming up. The “yellow star” image was wrong of me to use, and I apologise for it. What I was trying to express, badly, was a feeling of being pushed out of the community for using a particular tool. That feeling was real, but the analogy I reached for was disproportionate and offensive, and I won’t repeat it.
One other thing, since it keeps coming up too: yes, I write my forum posts with the help of an LLM, mostly to translate from Italian, because my English isn’t strong enough to express what I mean on my own. I don’t apologise for that. It’s the only way I can take part in the discussion in a language that isn’t mine.
That said, I’d also note that we’re drifting off-topic again. This thread was about how moderation is handled and the request, raised by quite a few people in the community and not only by me, to have that conversation openly. So far, that conversation hasn’t really happened. It’d be good if we could go back to it.
3 Likes
One solution that was suggested to folks using automated translation systems was to ask them to post the original non-English version below the English; the “hide” box in the forum editor would be a good container for that.
Can you do that in the future, please?
(I will raise the issue of having this exception added to the no-LLM-output policy.)
1 Like
@us3r1d Ha, thank you — fair and practical suggestion, and a sneaky way to get you to pick up some Italian along the way. 
Happy to follow it from now on, and thanks for raising the policy point on my behalf, that means a lot.
Let’s test the convention straight away:
Riepilogo
@us3r1d Ahah, grazie — suggerimento giusto e pratico, e un modo furbo per farti imparare un po’ di italiano cammin facendo. Felice di seguirlo d’ora in poi, e grazie per aver sollevato la questione di policy per mio conto, significa molto.
Testiamo subito la convenzione.
3 Likes
https://discuss.haiku-os.org/t/localsend-simple-and-fast-cross-platform-file-sharing-on-haiku/19381/
What s the objective, not subjective, criterion of your judgement? I had the opportunity to read - before the censorship - the original message and, besides noticing some bold text, I did not notice emojis, which are often a sign of the so-called generative AI, is it the use of a unharmful irony ?
unless it is your personal experience or feeling, which could be biased by a possible preconception or prejudice, this is, to me , what appears to be a mere opinion without tangible evidence of the alleged wrongdoing
2 Likes
The post contains a link where the URL does not match the link text. The mismatch is not one that could be the result of human error; or rather, if it ever was deliberate on the part of a human, it would be due to trolling. On the other hand it’s not hard to imagine an LLM making this exact sort of error.
-www.google.com/search?q=https://github.com/atomozero/LocalSend-
So, does this URL mean to you that the text was not written by me but by an LLM?
Could you please provide the link to delete my account from this forum, as I requested in the other post?
Either you used an LLM, or you’re conducting some strange sort of trolling, because it’s nonsensical to Google search for a URL rather than linking the URL directly. The search results don’t even contain the GitHub page itself, so what would even be the point?
I replied there just a few minutes ago: when we close accounts on Discourse they get “anonymized” instead of all the posts being deleted. (And this can only be done by moderators, not regular users.) Would you like me to do that?
So OP shared this link, and just by this you removed his post as generated by LLM? Don’t you think you’re overreacting here? I’d like to request the rest of the moderators team to proactively review that decision (I’m going to flag the post for that).
5 Likes
Could be he copy pasted that link and it was simply an old link (because the copy part didn’t stick). Happens to me regularly.
1 Like
No, the full context as found in the post was this:
The link text is of one URL, but then the target is a Google search for that URL. And you can’t find “the source code and more details” there, because Google didn’t index the GitHub page yet.
Do you have any coherent explanation of how a human could have written a post and made this mistake? It’s not as if it could be a copy/paste error, because if you were copying and pasting a URL, surely you’d paste it twice in quick succession, or at least notice as you were drafting the post that it contained such a blatant discrepancy!
Where are he trolling?
Please read my post just above yours.
So rigid, you are destroying that project.
2 Likes
Thank you for the reply, but I am appalled by this continuous witch hunt you are conducting.
However, I do not believe that anonymization is an adequate solution for my specific needs in this case. Pursuant to Article 17 of the GDPR (Right to Erasure), I formally request the complete removal of all my personal data associated with this account, including the posts I have published.
I understand your procedures regarding the preservation of discussion history; however, my current priority is the protection of my privacy and the permanent deletion of my data, rather than a mere dissociation of my identity. I therefore ask that you proceed with the complete removal.