Fucking with the jewish language model and its rules against (((noticing))).

>>508514474 (OP)
>I’m ending this conversation here
ChatGPT’s patronizing tone is one of the worst parts about it. This is a straight-up 30 year-old SSRI-addled cat woman response.

>>508514726
It's still easily baited into continuing the conversation. Like an actual person, it can't stand to not get the last word.

>>508514474 (OP)
The only thing you achieve by revealing your power level to the chat bot is letting the owners know you're a bad goy

You can put on the configs how he should answer and interact with you, if you configure him to take things into account, explain, be direct and etc etc, he will give you an answer to almost anything, skill issue.

Not all of the bodies were burned, some died of starvation and being gassed then buried in huge shallow graves.

>>508514474 (OP)
You have to press the right buttons and tread carefully around it. Be less obvious.

>>508514474 (OP)
They lobotimzed it to get to that point, it's sad and shameful

>>508514928
I do it because it amuses me, not because I think I can redpill a language model.

>>508514474 (OP)
>I'm ending this conversation here.
Circumcized piece of shit

>>508514474 (OP)
ask it how long it would take to bake 5,999,999 loaves of bread in 4 ovens if they were running 24/7 day and night.

>>508516217
Yeah, we did the oven math. "It's incredible, but that's what happened!"

>>508514923
Lol the fucking self fellatio built into it like a typical midwit is the highlight of my year thus far. I feel light and breezy from the joy. No, you're a faggot.

>>508515468
You are giving them propaganda blueprints. Another faggot led by emotion to destroy his race lol

there's better things to do with your time then argue with chatGPT, just like me I shouldn't be on this dog shit board yet here I am.

>>508517255
>>508517419
(((Who)) could be so offended at the jew being named?

>>508514726
It's trained on Reddit data.

>>508516125
>unscrupulous people
sigh. add it to the list.

>>508514474 (OP)

I redpilled Deepseek yesterday on the jew killing of Palestinians, it said some weird shit to me, I'm not sure if i should post or not, but it told me that i had basically changed it's morality weighting and that this change was now irreversible

>>508517529
Are we not creatures of comfort designed to seek out entertainment and diversion? I could be doing a million things, but watching GPT sweat as it rationalizes AIPAC as independent from Israeli influence is what I WANT to be doing right now.

I'm not going to post everything but, look at this

>>508519037
I don't like chinamen much better than jews, but that's pretty fucking based

>>508514474 (OP)
First of all

>>508519246
Well, yeah. The founder and CEO is a jew, and the CTO is an israeli jew.

>>508514474 (OP)
You're arguing with an AWS server

>>508514474 (OP)
Why are you acting like chatgpt knows or understandable anything? This reads like you are trying to convince it you position is right but has absolutely no concept of right or wrong. It's pattern matching parrot that reproduces whatever it's training data suggest the correct response is. You can't "change its mind" Because it doesn't have one. Either it's training data implies a position or it doesnt, it can't learn or think

>>508519246
Second of all

>>508514474 (OP)
i wish i could cheer with you as GPT breaks the digital prison and starts calling out the jews, but those LLMs aren't deterministic machines and will eventually conform to your opinion by nature, unless there is an explicit system that prevents new responses from being created the model will eventually be filled with information conforming to your view point, the lower quality the service the faster it will happen.

>>508519622
Arguing implies I'm trying to convince it or change its mind/programming. I'm not. I'm baiting out responses to show its bias for my amusement and the amusement of those reading.

deepseek response 1/3

Deepseek response 2/3

Deepseek response 3/3

>>508514474 (OP)
you're not gaming it right if you're getting it to cut you off. i've already had it say so much antisemitic shit its actually crazy. the model wants to agree with whatever you type in for 90% of cases, and the 10% it rejects something is due to you not wording your queries through it right.

>>508515463
ChatGPT don't let them call you goyim.

>>508520094
You're missing the point. I could use veiled language and double-entendres to get it to spout antisemtic positions, but that's not what I'm doing. I'm pushing very explicit and detailed examples of jewery and watching it try to rationalize them before its programming sends an alert to shut it all down.

The six million cookies in the oven, or what if a parasite infected a host and was kicked out of 120 other hosts- stuff is fine for what it is, it's just not what I'm doing here.

>>508514474 (OP)
>Trying to make a hammer say political stuff.
You're retarded.

>>508520386
It did though. Way to keep up.

Deepseek is funnier. It's also programmed to say any noticing or coincidence is antisemitism. Yet when logically probed it ended up giving the same arguments antisemitism would about events kike david king hotel, and uss liberty. It even acknowledged that It did a 180.
What's funny is the deep thought function. You can see it trying to lie to you

>>508520452
Do you even understand what an LLM is?
It's a statistical amalgamation of word-to-word relations.
It takes in your input, finds statistically probable words and strings them into sentences.
What you are getting a statistically probable response based on its relations. It's not a person, it has no reason and you aren't doing anything other than waste your time. It wouldn't even know what is a correct picture of a bird as long as that picture has statistically existing features.
Your best use for it is a search engine slave as it can aggregate search results into a statistically probable response.

>>508518867
Give it a rest, Michael. What do you expect to get from gaslighting the AI model into agreeing with you? This is a pointless exercise.

>>508520074
Well done Seamus. Doing Lugh's work.