Proof of AI-assisted political profiling by Unruffled @ lemmy.dbzer0.com
-
from the mod log:
The output from the LLM was provided as proof that someone needed to be banned.
I didn't want to do this but my hand was forced by their dissembling, minimization and bullshit. Note how the output includes a nice big logo with a slogan and ascii graphics - this was not a one-off experiment, this is an app they were proud to call their own.
Well...First, i want to thank you to you all, the topic wasn't easy but the discussion is going well.
I hope the thread will remain that way :)
-
Instead of telling myself and others that we couldn't possibly be anarchists because our view on the subject are different to yours and suggesting someone read a 270 page book to understand your POV, why don't you summarise your reasoning?
Is it that AI is able to free up our menial labour?
Is it that is infringes on corporate copyright and steals from them?
Is it that it's being used in a manner that harms the finances of AI companies who charge for this service?
Instead of telling myself and others that we couldn’t possibly be anarchists because our view on the subject are different to yours and suggesting someone read a 270 page book, why don’t you summarise your reasoning?
Have you noted the many words I put in italics in previous reply? Maybe there was a reason I insisted so heavily on those. And, once again, don't waste your time (and mine) trying to put words in my mouth that I've not said. The only reaction this might end up triggering (you've got some leeway) is in me deciding to ignore all your messages from that point on.
As for why I don't summarize the book for you and expect you to "read a 270 pages book"? Thx a lot for asking.
We live in the age of "summaries". People, old and young, don't want to read anymore. They want an instant summary of those books so they feel like they know it without having to actually put any effort reading it. So they feel confident they can get good grades when there is an exam on that book (the only ones to blame here being the educative systems and the teachers, not the kids abusing the system as they're acting quite smart by doing so). Or feel like they can knowingly talk about that book with their peers.
People who want to use that shortcut are more than welcome to go that road if that's what they really want, you are more than welcome to go find yourself a summary of the book, even though I think it's a sad mistake. You can even ask one of the many AI to do summarize it to you (maybe this could give you enough motivation to start reading it?)Just don't expect me to summarize the book for you for the simple reasons that I think one must be willing to read a book not just act like if they had (it's sad but it's ok to not read at all), and then because I think most of the value one can get out of reading lies in the act of reading the actual book, taking the required time to read it and not in the act of getting a ready to memorize summary of it.
Instead, what I'm willing to do, if you ever decide to read it and put in the work, it is to take time to discuss it at length with you, discuss any question or objection you might raise... to the best of my own limited abilities.
Obviously, that is just my opinion, feel free to ignore it.
-
Same, i will probably lock it during my sleep time than unlock when i'm awake. I did the same with a luanti server.
For now the thread is going well. Thank to them all. :)
And don't do mod task if you only wanted to do the wiki or something else. ;)
And don’t do mod task if you only wanted to do the wiki or something else. ;)
Oh don't worry about that ;)
It just happened I considered the OP post oddly interesting to comment on it (even though I still think this is not the best place to discuss that kind of questions, it also opened a few very interesting perspectives worth exploring) and I also wanted to make it clear to the person that 'reported it that it had been dully noted by a mod (ie me wearing my mod hat, in my first reply) and that said mod saw no real issue in the OP post.
After that, the few feedbacks I received had promises in them. I tried to reply as openly and as usefully as I could. Alas, it seems we're quickly moving towards a more ad hominem type of discussion around the classical "you're doing it wrong or you must be dishonest if you don't agree with us/me" type of exchange that I won't contribute much more to if that doesn't change. Will see.
-
And don’t do mod task if you only wanted to do the wiki or something else. ;)
Oh don't worry about that ;)
It just happened I considered the OP post oddly interesting to comment on it (even though I still think this is not the best place to discuss that kind of questions, it also opened a few very interesting perspectives worth exploring) and I also wanted to make it clear to the person that 'reported it that it had been dully noted by a mod (ie me wearing my mod hat, in my first reply) and that said mod saw no real issue in the OP post.
After that, the few feedbacks I received had promises in them. I tried to reply as openly and as usefully as I could. Alas, it seems we're quickly moving towards a more ad hominem type of discussion around the classical "you're doing it wrong or you must be dishonest if you don't agree with us/me" type of exchange that I won't contribute much more to if that doesn't change. Will see.
Well don't hesitate to lock with a message that explain why you lock the post if we move toward this kind of exchange.
It can excalate quickly and we have lot things to do. :)
-
Well don't hesitate to lock with a message that explain why you lock the post if we move toward this kind of exchange.
It can excalate quickly and we have lot things to do. :)
Will do if I have to, but like I told the person I'm more than willing to give them an opportunity to keep this a real discussion, not an ideological pillow fight ;)
If you read it, so far it is still perfectly correct with barely a few hints at... anger, or not even that: a mere desire to trigger some emotional reaction on my part maybe.
-
I've not followed along all the various threads and, in all honesty, I've only read this very post because it popped up as being reported while I was trying to drink my morning coffee. Also, don't ask me how I ended becoming a mod either, I have no clue.
Imho, and I say this as someone only minding the meaning of words, if those people really see themselves as anarchists I would suggest they urgently:
- Look up the definition of anarchism in a (decent) dictionary and see how well it fares using tools like AI (being a centralized tool that is owned by some powerful/uncontrolled group, the IA itself being a true black box... while they, the 'anarchist' users of said AI themselves trust it, and so on). Or maybe, since they're already using it, they could ask their AI if they're really meant to go hand in hand? ;)
- Spend some time finding a better suited name to describe their jolly band. Imho, "believers" seems much more suited than "anarchists" to describe any group of people willing to listen to words coming out of some black box outside of their control but that is just an impression, there may be better words out there.
I don't mind people using AI: it's a new tool and it is to be expected that some real stupid shit will be done using it. But, like any tool, AI is not neutral. By and in its very design AI has a purpose and if I was to define myself as an anarchist, which I would not if there was still any doubt about that, I would seriously and very urgently question my desire to use AI despite its intrinsic design, despite its very purpose.
Now, if you don't mind, I will pour myself a fresh and hot new cup of coffee. And maybe later today I will pick up my copy of Proudhon 'Qu'est-ce que la propriété?' ('What is Property', which was first written in French) which, imvho, any wannabe anarchist should read before deciding it's ok, or not ok, for them to use AI, and what type of AI. And yes, that is a book worth reading for discussing AI despite having been written in the first half of the 19th century. The answers they might get out of that book may very well... unsettle them and their apparently very unchallenged certainties.
A very stimulating and healthy read, even for non-anarchists like myself.
"I don't know anything about the situation at hand, the people in question, nor the political theory they espouse, but I won't let that stop me from stating my opinions as facts with the confidence of a hundred cardinals. Also, read this 200-year-old political treatise book that will definitely apply to bleeding-edge GenAI technology."
-
"I don't know anything about the situation at hand, the people in question, nor the political theory they espouse, but I won't let that stop me from stating my opinions as facts with the confidence of a hundred cardinals. Also, read this 200-year-old political treatise book that will definitely apply to bleeding-edge GenAI technology."
Almost Impressive demonstration of proud(hon) ignorance. Enough with that.
-
from the mod log:
The output from the LLM was provided as proof that someone needed to be banned.
I didn't want to do this but my hand was forced by their dissembling, minimization and bullshit. Note how the output includes a nice big logo with a slogan and ascii graphics - this was not a one-off experiment, this is an app they were proud to call their own.
Dbzer0 is full tilt groupthink and exclusionism. They consistently follow their leader and error-prone tech to aggro any form of defection— it’s hard to parse it from a cult. Rimu stays based for calling out them assuming guilty before proven innocent. I only wish the Piefed full instance block feature still worked on the Voyager app.
-
This is a whole lot of absolutely nothing. Thats why.
BREAKING NEWS! Mod team develops and uses mod tool!
„Anarchist” mod team delegating work to a non-deterministic tool operated by big tech is the news here. Kind of confirms my understanding of who calls themselves anarchists these days - people who value freedom to consume and freedom to be exploited above actual freedom.
-
from the mod log:
The output from the LLM was provided as proof that someone needed to be banned.
I didn't want to do this but my hand was forced by their dissembling, minimization and bullshit. Note how the output includes a nice big logo with a slogan and ascii graphics - this was not a one-off experiment, this is an app they were proud to call their own.
Unruffled also doesn't tag AI posts, violating db0's own AI policy. But db0 doesn't seem to care. The place is getting more and more obnoxious by the day it seems.
-
„Anarchist” mod team delegating work to a non-deterministic tool operated by big tech is the news here. Kind of confirms my understanding of who calls themselves anarchists these days - people who value freedom to consume and freedom to be exploited above actual freedom.
There is no delegation to the tool.
It provides an evaluation in text format, and references. Thats it.
Edit: To be clear, rimu is just making that part up. It has no ability to auto-ban and its not an automod.
Its a python script to check a user's comment history and evaluate using a local model, and spits out what it finds. A person has to read it and make a decision on this information.
Everything else is rimu's hallucination.
-
And? It's in the public modlog, Rimu. Please, can you stop harassing me? How many hater posts are you gonna make this week exactly?
this was not a one-off experiment, this is an app they were proud to call their own.
It's a python script that calls the lemmy/piefed api to grab comment histories. Calling it an "App" is quite a stretch. Luminous just likes pretty things, bless his cotton socks.
You are right about one thing, we will no doubt keep using the script to pull comment histories on the few occasions when we want a deep dive into someone's profile.
For the record, we were never planning to use it as an automod or hook it to an LLM. You hallucinated that, ironically. That's why we never announced it. It was nobody else's business.
How it was actually intended to be used was as a tool to supplement the existing lemmy mod tooling, so that mods can make better, more informed decisions, in a more efficient way. How is it ever a bad thing to have all the relevant information to hand? And it's being open sourced soon, so everyone can use it. Unlike some instances [nudge, nudge], we try to take a considered approach to user bans. And no, it's no linked to an LLM in any way. If you chose to use one, you'd have to paste in the user history manually. But it also very useful just for skimming through, much faster than with Lemmy UI.
Also for the record, we do not use any form of automod with automated banning and never have. All ban decisions are made by a human mod or admin. Neither do we auto-scan user profiles for bad political takes, like Zionism. What we do do, is act on user reports, conduct an investigation and then make a mod decision. Nothing has changed in that respect, except now we have an extra tool to make sure we get the decision correct.
Finally, just because I included an LLM summary of the comment history in the public modlog, it does not logically follow that the ban occurred because of the LLM summary. It simply happened to correspond very well to my own manual review, which I have already explained to you. And I thought the summary did a much more comprehensive justification of the ban than I could have fitted in the modlog, so why not include it?
In conclusion, much ado about absolutely nothing. Again.
Aside from all the drama, is there a reason you are using the gpt-5-mini api from OpenAI (and aparently pay them for that) instead of a self-hosted model? This seems to be also against your instance policy and helps OpenAI to harvest training data.
-
Aside from all the drama, is there a reason you are using the gpt-5-mini api from OpenAI (and aparently pay them for that) instead of a self-hosted model? This seems to be also against your instance policy and helps OpenAI to harvest training data.
We added the chatgpt reference because we thought it would be funny, luminous talked about it somewhere already. We are pirates, and self-hosters. This is surely obvious. And we don't have cash to splash around on corporate AI tokens.
-
Also, don’t ask me how I ended becoming a mod either, I have no clue.
Ah you don't remember but it was for editing the piefed wiki and the fedihack event. The rest, i completely forgot.
Considering the recent actions, maybe reconsider that.
-
We added the chatgpt reference because we thought it would be funny, luminous talked about it somewhere already. We are pirates, and self-hosters. This is surely obvious. And we don't have cash to splash around on corporate AI tokens.
So what model are you using? Models are biased by their training data and guardrails put in by the trainers, so for transparency sake it is important to mention the model used.
-
I think it's notable that the big receipt reveal is an example of the LLM mod tool absolutely nailing the assessment.
Honestly - this type of reporting for ban rationals should be the gold standard of moderation, AI or no. If an AI tool can help raise the standard for transparent moderation, I am 100% in support of it.
Civilian de‑protection rhetoric
Notable examples:- Claims that “minors are not innocent by default”
- Statements that “they’re killing children” is mere outrage propaganda
- Questions implying all Palestinian deaths may involve combatants
Yeah I'd like to see direct quotes but even if this is heavily paraphrased there's no question.
-
Civilian de‑protection rhetoric
Notable examples:- Claims that “minors are not innocent by default”
- Statements that “they’re killing children” is mere outrage propaganda
- Questions implying all Palestinian deaths may involve combatants
Yeah I'd like to see direct quotes but even if this is heavily paraphrased there's no question.
Full comments with direct links are there, thats the summary up top, just scroll down.
Edit: https://faf-pb.xyz/?ab3c4aae88b64dbc#J8jz6mNac8kRHm5aMXBJQjbUr9pbwMfWrdEaNm1jywhx if you want to see for yourself.
-
We added the chatgpt reference because we thought it would be funny, luminous talked about it somewhere already. We are pirates, and self-hosters. This is surely obvious. And we don't have cash to splash around on corporate AI tokens.
Sadly this is a learning moment that inside-jokes don't belong in modlogs. Any nuance it would have would be immediately lost and twisted beyond any recognition.
-
So what model are you using? Models are biased by their training data and guardrails put in by the trainers, so for transparency sake it is important to mention the model used.
Qwen iirc.
-
"I don't know anything about the situation at hand, the people in question, nor the political theory they espouse, but I won't let that stop me from stating my opinions as facts with the confidence of a hundred cardinals. Also, read this 200-year-old political treatise book that will definitely apply to bleeding-edge GenAI technology."
Accidentally unlocked the comment, but I don't see why it should be locked in the first place.
Re-lock if it's supposed to be locked, I guess.
Ciao! Sembra che tu sia interessato a questa conversazione, ma non hai ancora un account.
Stanco di dover scorrere gli stessi post a ogni visita? Quando registri un account, tornerai sempre esattamente dove eri rimasto e potrai scegliere di essere avvisato delle nuove risposte (tramite email o notifica push). Potrai anche salvare segnalibri e votare i post per mostrare il tuo apprezzamento agli altri membri della comunità.
Con il tuo contributo, questo post potrebbe essere ancora migliore 💗
Registrati Accedi