The absolute state of chess dot com censorship

I like threads like this, from a list of restricted words to Orwell and matriarchy...

You guys can read Teslas thoughts here, I'll post an excerpt or two.

"It is clear to any trained observer, and even to the sociologically untrained, that a new attitude toward sex discrimination has come over the world through the centuries, receiving an abrupt stimulus just before and after the World War.

This struggle of the human female toward sex equality will end in a new sex order, with the female as superior. The modern woman, who anticipates in merely superficial phenomena the advancement of her sex, is but a surface symptom of something deeper and more potent fermenting in the bosom of the race.

It is not in the shallow physical imitation of men that women will assert first their equality and later their superiority, but in the awakening of the intellect of women."

Better pucker up guys if you don't want to be a victim!

I m sry but overestimating someones capacity to be better or worse on anything based just on his/her sex is still sexim, women have proved to be equally cruel and bad as men on many areas.

I believe more in the individual no matter their sex, because the most crucial factors on someones development is his/her gene pool, his family and the environment he she grows up.

A good range of related topics 😅🙈

If WE are evolving into a matriarchy, then what are WE now?

Nothing you said has anything to do with an AI chatbot's ability to determine context in terms of filtering for bad language. I will also note that you as a human being only know what words *can* mean yourself, not what a speaker may intend. So your ability to judge what words mean in context is not 100%, like an AI chatbot's is not 100%.

P.S. Biological imperatives are not restricted to the desire to have children, nor are they restricted to one sex. Off the top of my head:

- Desire to reproduce and protect offspring

- Hunger/thirst

- Reaction to drowning/burning/temperature extremes

- Getting sleepy

- Reflexive pain avoidance/withdrawal

- Reaction to being confined/unable to move/escape

- Revulsion based on certain sensory input

Etc.

yes its an inverted form of sexism and should probably be illegal.

Somewhere in between

The great Fischer, peace be upon him

Interviewer : A woman's place is over the hot stove, eh (rhetorical question)

Bobby 2 Awesome : No they cannot cook!

Bobby is of course talking pants but its funny never the less, he forgot about Vera Mencikova and later praised Nona Gaprindashvili

Diogenes, your arguments are verging toward the judgment and assessment of machine intelligence by comparative measures. It's a deep hole that you're flirting with, and this too is fascinating.

I've heard by more than a few the claim that LLMs have "crushed the Turing Test." This is an absurd claim on its face, but I stand with those who have yet to see a persuasive demonstration of machine intelligence even in narrow, specialist domains.

As to my knowing what you are thinking, yes, yes, I can only infer. And maybe I get it 50-50 or less even with my best friends, as do they

But I know that you are not a mere automaton, and I know that the color blue looks blue to you, and doesn't sound or taste blue. And I know without a doubt that you dream, Diogenes. How? Because I know that I do.

Machines have a lonnnnng way to go, and thrir fandom trusts them with too much.

Well I don't think THIS post will get censored but I don't know WHERE to post it. If you do,....please transfer. Thank U. https://sports.yahoo.com/article/harder-losing-basketball-nba-legend-101103184.html

More hand-waving. and dodging...the Turing Test toss-out is a straw man.

LLMs can easily distinguish contextual usage of words and phrases. Here's the proof:

"ChatGPT, what do you think of this argument [your post inserted] against LLMs eventually being used to provide forum moderation? Can LLMs filter out "bad" language contextually?"

As an argument against LLMs for forum moderation, it’s on shaky ground. Here’s why:

1. Can LLMs filter “bad” language contextually?

Yes, technically.

Modern LLMs already do context-sensitive moderation — they can distinguish “---- you” (insult) from “---- this game” (frustration) from “---- it, I’ll try again” (self-directed). They can also handle nuance like “this is stupid” (mild criticism) vs “you’re stupid” (personal attack).

Limits remain.They’re still brittle around sarcasm, coded language, regional slang, dogwhistles, and in-group banter. A human instantly spots when “nice job, genius” is hostile; an LLM might misclassify it or let it slide. 

2. The “fundamental flaw” in the argument: If someone says “LLMs can’t be used for moderation because they don’t understand”, that’s a category mistake. Moderation is pattern recognition + policy enforcement. It doesn’t require true understanding, only consistency within a rule set.

LLMs are already better than regex filters (which forums used for decades).They still need human review for appeals and edge cases. 

3. Passing the Turing Test and doing context filtering are very different problems:

Turing Test ≠ moderation competence:

Turing Test goal: fool a human into believing they’re talking to another human across open-ended conversation. That requires broad generality, personality simulation, and sustained illusion.Moderation goal: detect whether text violates a narrow set of rules (e.g., insults, harassment, p-rn, spam). That’s classification, not conversation.You can have a model fail miserably at the Turing Test yet still be excellent at flagging “go kill yourself” as harassment.

Context ≠ consciousness:

Context filtering requires statistical inference on usage patterns.Example: A system doesn’t need to “understand” why “genius” is sarcastic; it just needs enough data where “nice job, genius” co-occurs with negative sentiment and insults.This is why sentiment analysis, toxicity classifiers, and content moderation tools (Perspective API, OpenAI’s moderation models) already outperform dumb keyword bans — no Turing Test mastery required.

Gee, the LLM seems to do just fine breaking down your post contextually...it also identifies the things it cannot always distinguish, which fall into the same categories as things humans cannot always distinguish.

Your last summary statement is silly. LLMs do not have a "long way to go" to provide better filtering than chess.com employs now. That's painfully obvious, regardless of your attempts to turn this into an "AI Pinocchio will never be a real boy..." discussion. You're trying to sound intelligent and dismissive here, but it's not working. You should have picked a better hill to die on.

More censorship making the site unworkable.

Attempted to post this hyperlink to a scholarly article,

https://moderndiplomacy.eu/2026/03/05/the-missile-saturation-gambit -chinas -strategy-to-lure-american-and-israeli-firepower-into-a-costly-trap/

but could not because apparently gambit china cannot be used together. Seriously?

Ah, memories unlocked (September 2025)

the censorship on this site is still quite absurd, if I do say so myself.

It is still not a good reason to create irrelevant comments in threads of other users

there is a forum for advertising, move your pollution elsewhere.