Maybe I'm wrong about Claude/Anthropic, but my understanding is most of these models are trained on data publicly available on the internet. Sure they paid Microsoft for access to your git repos/Reddit for your posts and comments, but they don't own that data*. Copyright on AI generated content is still in a pretty grey area legally, so I don't see what they can do about people scraping their output except for terminating accounts for TOS violations, especially since I think everyone they're calling out is based in China.
Ultimately, if this is actually a threat to their business model (I know, I know, their business model is fraudulent and incestuous, not based at all on paying customers) they should release their own distillations. If you're going to base your models on publicly available data, and have a public API for accessing it, its outputs are effectively in the public domain, whether you like it or not.
*They license it from you with pretty broad terms on what they can do with it, but it's still your data.
"If God didn't want [x] he'd just make it impossible"
Single-digit IQ take.
Not racist enough. Skyrim is for the Nords!
Especially when he's posting your private communications so publicly.
"Women were samurai, except none of them were warriors, and we won't show you any pictures of female samurai."
Summary of the article for those who have trouble with the archive or don't want to click through.
He sold out on his last book I believe, gay characters or some shit I forget.
He put gay characters in the WoT books he (co-)wrote. And he made sure to explicitly point out that the fact that it was OK that this minor character who only played a role for a few chapters in the last few books was gay.
I don't recall Jordan doing anything other than making sideways allusions to any characters other than the Forsaken (several of whom were complete sexual degenerates) being gay.
the FBI were actively surveilling the riots (as we know from that 3rd party contractor's drone flight that was presented to the prosecution in Kyle Rittenhouse's case in violation of discovery rules).
As an aside, the wild part about that is that the drone took off seconds before the child molester started chasing Kyle.
You're gonna catch flak for keeping the fag's mod's Title and TLDR.
Don't you know that stuff is toxic?
Leftism benefits women (and other 'protected groups') in the short to medium term, but ultimately degrades quality of life as the system collapses under its own weight. Conservatism (better phrased as "western traditional values", as no ideologies called "conservatism" have conserved anything) elevates those capable of leading, resulting in a better quality of life for most people over a medium to long term.
Trafficking minors (under 18) across state lines for the purpose of engaging in sexual activity is still a federal felony even if they're above the age of consent in both states.
If my memory is correct, $1,000,000/year will be a middle class salary by 2065. Now, that's the salary of someone about to retire, and assumes that wages keep up with inflation (which they won't), but things are going to get really stupid in the job market over the next few decades.
Name ONE Democratic president that ok'd a successful military operation
Andrew Jackson. Which doesn't really contradict your point, but...
Not to burst your bubble or anything, but this guy's been around for like 20 years, so stuff this funny has been around for like 20 years, just (as you point out) 'the algorithm' didn't share it with you.
Anyway, remember to stomp the groin.
This cop is scum because other cops shot people? Did I read that right?
Peggy
That face
Something tells me this one might have a little extra downstairs.
I'd say ask the admins what triggered it, or DoM, but I doubt either is going to give you a response. Without knowing how Scored moderation works, my guess is as good as yours, but I'd agree that it's almost certain that this was some sort of auto-mod triggered by keywords or somesuch.
I'm just gonna go through the detailed rules from the stickied "Welcome Ashore!" post.
I've created a list of rules as below, they
will likelywon't change later unless the whole site dies down and I have to give up moderating duties, but they are here for the purposes of establishing a base level of social order:
ONE: Do not post Illegal Activity. Also, do not post any manifesto's done by terrorists, active shooters, serial felons rationalizing such things, or promoting such things, even if your content does not endorse the message.
Unless this is a sitewide rule, it should be removed. The value of being able to analyze and discuss the trannifesto or "The industrial Revolution and it's Consequences" outweigh any potential benefits of censorship.
TWO: Do not engage in speech that promotes, advocates, glorifies, or endorses violence.
Considering that this has literally been used to delete posts advocating for capital punishment, it should also be removed, or at least pared back to require an imminent and specific threat of violence.
THREE: Do not threaten, harass, or bully users; and do not encourage others to do so on or off-line; nor make per se defamatory states at users.
I'd rephrase this as "Don't engage in criminal harassment of other users" and leave it at that, otherwise the rule is intentionally overbroad.
FOUR: Do not post ISM. Involuntary Salacious Material means NSFW material of a manner that was not intentionally made public. This is the "upskirt", "revenge porn", and "private intimate photos" rule.
FIVE: Do not post Porn
Probably the first two rules that are fine as-is.
SIX: Content that contains nudity, pornography, or profanity, which a reasonable viewer may not want to be seen accessing in a public or formal setting such as in a workplace should be tagged as NSFW. Any material of a titillating nature must be marked NSFW.
This needs to be rephrased, since "porn" is already explicitly banned, but otherwise I think it's OK.
SEVEN: Do not post Facebook accounts, individuals who's twitters are less than 500 followers, private/personal information that is not publicly available, addresses, or participate, encourage, or engage in any doxxing campaign.
The first half of this is fine, though '500 followers' is a bit arbitrary. The second half should be kept inline with sitewide rules, but otherwise shouldn't be needed.
EIGHT: Do not intentionally deceive others by impersonating another. This does not apply to satire.
I think the spirit of this rule is fine, and I don't think it was ever abused in practice, so I guess we can keep it.
NINE: No person shall use communities.win sites (including kotakuinaction2.win) to solicit, facilitate any transaction, or gift including: ... ATF defined firearms or ammo as defined by the ATF, Bump-stock type devices, Explosives, 3D printing files to produce the aforementioned, controlled substances, Drugs, Alcohol, Tobacco, Stolen goods, Paid services involving physical sexual contact, Personal Information, Falsified Official Documents, Falsified Currency, Fraudulent Services, Pharmaceuticals
This looks like a direct restatement of a sitewide rule.
TEN: No vote manipulation. Do not break communities.win's features.
Probably also a direct restatement of a sitewide rule, but if it isn't, I'm fine keeping it.
ELEVEN: Do not post spam. If you are self-advertising, you must have sufficiently engaged in the sub prior to your post, and you must engage with the users when they comment in your post. Spam will also include repeated messages and comments that are done with no effort to add to the conversation.
I don't recall this rule ever causing issues, and there are certainly a few users that would have caused issues if they weren't limited to 5 posts/day, so I'd say we keep it.
TWELVE: Do not post intentional falsehoods or hoaxes. Yes, the Elders of Zion and other such intentionally fabricated documents fall into this. If your POST is arguably false by the user-base, it may be marked as either misleading or unfounded based on it's factual assertions, particularly in the title.
I don't like this rule, because it rarely seemed to be used to remove intentionally misleading content, as opposed to just content that one of the mods didn't like. I say we remove it and, if necessary, replace it with a re-worded 'misinformation' rule.
THIRTEEN: If you have reposted something, it will be removed
This needs to be adjusted, because the same content will often be posted twice (rarely more) in the same day/week, and the second/subsequent posts will attract more discussion due to timing/title/whatever. We should keep some form of this rule, but enforcement should take care to not quash discussion.
FOURTEEN: Do not post more than 5 posts a day to this sub.
Oh, this is the actual anti-spam rule I've seen applied. We can keep both, I guess.
FIFTEEN: Do not direct particularly egregious identity based slurs at users. A list will be provided
SIXTEEN: Do not attack entire identity groups as inferior, subhuman, inherently morally deficient, biologically/evolutionary mongrel, or participating in a vast conspiracy to take over the world, ala ZOG-NWO / The Patriarchy.
Fuck right off with this shit, you kike faggots.
I know it really incenses you when someone quotes your own magic spells back at you and robs them of their power, so allow me to provide context:
[The people of the UK] don't have guns, that's why [they accept mandatory digital ID]. [...]
Admitted brown paid jewish shill account lies randomly about the efficacy of guns. Claims the things that have not been used since 1774 [emphasis mine] will magically be used, ever, to stop anything that is happening now, despite no evidence thereof. More at 11.
It is required by its contract of employment to reply to this comment, otherwise it will be fired.
In conclusion: fuck off, faggot.
"Because I work for Ben Shapiro and I dont want to be next."
This stuff was first posted here years before Kirk was assassinated. That twitter post is from 2021.
Hey faggot, how about you try not reposting (with the same exact title) a post you already made, just because you didn't like how people responded to it the first time?
https://kotakuinaction2.win/p/1ARd8YdYdL/a-judge-in-the-netherlands-sente/c/
And watch, he'll delete this one too, now that two people have pointed it out.
That show got 6 seasons? I guess that was before Hollywoke had started making nothing but flops so they could keep an unprofitable show going for that long, but still...
This is YEARS old, and posting a screenshot without a date on it is incredibly misleading.
Do you know how LLMs work? There's no 'logic', it's all statistical modeling. What DeepSeek et. al. are doing is taking inputs from specific benchmarks, throwing them into Claude, then training their (smaller, cheaper, faster) models to match the output, bypassing the need to scrape the broader internet (or pay) for training data. Which is (IMO) why Anthropic is actually salty; they paid for publicly available data (to avoid being sued) and someone outside the legal system bypassed that process.
I guess you could call that 'backing out' or 'reverse engineering', except none of these companies actually understands the inner workings of their models (they have billions of parameters and are just too complicated), just the processes used to create them. It's a black box full of linear algebra.