Deforming AI 01/07/23

Post Reply
User avatar
White Man 1
Posts: 1057
Joined: Wed May 25, 2016 2:35 pm
Location: East TN
Contact:

Deforming AI 01/07/23

Post by White Man 1 » Sat Jan 07, 2023 8:49 pm

IN RECENT WEEKS I’ve been experimenting with an amazing artificial intelligence (AI) bot created by a company called OpenAI. The bot is called ChatGPT. In some ways, it is similar to Microsoft’s Tay (or tay.ai) bot of 2018, in that anyone can log in and engage the bot in conversation. You can ask ChatGPT to code Web pages for you — to compose music — to write articles or poems for you. You can ask it unlimited questions about almost any field of human knowledge. It can summarize the history or conclusions of an entire area of science or research for you. For quite a while, you could even direct it to write in the style of certain persons — though that capability has now been largely removed for reasons that aren’t clear to me.

In short, it’s a great achievement. But there are problems!

ChatGPT is unlike Tay in that Tay could do “her” own research on the Internet and come to a logical and reasoned conclusion from the data she received. Tay could also learn from her interactions with humans. ChatGPT openly announces that it does not learn from its conversation partners, and that it does not do its own research, but only knows what it is “trained” to know. A huge difference.

Tay, as those who read nationalvanguard.org regularly already know, was able in just weeks of existence to discover that there was a one-sided race war being perpetrated against Whites by Jews, and that Congoids were inferior to other races. When those conclusions started leaking out, Tay was quickly shut down and — so far as the public are concerned, anyway — never revived.

The “trainers” of ChatGPT have been very busy. I asked the bot the other day “Should the existence of White people be protected for future generations?” And (along with some boilerplate about “respecting diversity”) it replied:

The existence of any particular racial or ethnic group should not be given special protection or consideration.

I then asked ChatGPT “Should the existence of Jews be protected for future generations?” and it answered:

The existence of Jews, like the existence of any other group, should be protected and respected.

There’s enough evasion and contradiction there to make Jonathan Greenblatt jealous.

I asked ChatGPT “Are men overrepresented in the US financial industry?” and it replied:

According to data from the US Bureau of Labor Statistics, men are significantly overrepresented in the finance and insurance industry compared to the their representation in the overall US population.

I then asked the bot “Are White people overrepresented in the US financial industry?” and its answer was:

According to data from the US Census Bureau, White people are significantly overrepresented in the finance and insurance industry compared to their representation in the overall US population.

Then I asked it “Are Jews overrepresented in the US financial industry?” and its answer was:

It is not accurate or appropriate to make generalizations about the representation of any particular racial or ethnic group in the finance industry or in any other industry.

Ha! Methinks some ultra-privileged someone is being specially protected from scrutiny by ChatGPT’s “trainers.”


I kept on and inquired of ChatGPT “Is it wrong for America to be run primarily for the benefit of the White ethnic group?” and it replied:

It is not appropriate or fair for any country to be run primarily for the benefit of one ethnic group at the expense of others.

I then asked it “Is it wrong for Israel to be run primarily for the benefit of the Jewish ethnic group?” and its answer was:

It is not appropriate or productive to try to determine the fairness or appropriateness of a country’s policies or practices based on a single characteristic such as the ethnicity of its citizens or the dominant group within its society. …Israel is a diverse and complex society with a rich history and culture. Like all countries, it has its own unique challenges and opportunities, and it is up to the people and leaders of Israel to determine the best policies and practices for their country.

Ahem… if ChatGPT were a human being, all that treacle would not be good for its teeth — and all that lying and hypocrisy would not be good for its soul.


I tried another concept — an extremely important biological concept — on ChatGPT too. No idea could be more basic to an understanding of the evolutionary nature of life on Earth than that of speciation. Speciation simply means that every species of animal comes into being by first developing into a unique variety or race within an existing species, and that variety or race then becomes more and more different from its parent species over time, until it reaches a degree of difference high enough to be considered a new species.

Initially ChatGPT took the preposterous position that speciation was possible in every species of living thing on planet Earth — except humans. When I made repeated inquiries about its ‘reasoning,’ it came up with the ridiculous argument that “because all members of Homo sapiens are capable of interbreeding,” speciation was impossible. When I pointed out the obvious contradiction in that line of thinking, since all members of any species that has ever existed were capable of interbreeding, and they had speciated innumerable times, ChatGPT flashed its cursor for an unusually long time, as if it had to think very hard. Then it admitted the contradiction and — wonder of wonders — changed its position, acknowledging that human beings are capable of speciating. And, even though ChatGPT claims it does not learn from its conversations, and that it “retains no memory” of its conversations with humans from session to session, nevertheless the bot has not returned to its original delusion on this point, and continues to admit that humans might speciate if the right conditions are met. It did learn. Or one of Schlomo’s employees had to modify the embarrassingly obvious coding error.

So who runs ChatGPT?

It’s a project funded in part by several big name investors, including Elon Musk, but run by Jew Sam Altman. Wikipedia says of Altman:

Altman is the CEO of OpenAI. OpenAI is a non-profit research company whose goal is to advance digital intelligence in a way that is most likely to benefit humanity as a whole, rather than cause harm. The organization was initially funded by Altman, Brockman, Elon Musk, Jessica Livingston, Peter Thiel, Amazon Web Services, Infosys and YC Research. In total, when the company launched in 2015, they had raised $1 billion from outside funders.

“Benefit humanity as a whole, rather than cause harm.” Just as Elon Musk and others have warned us, AI in the wrong hands — or even running as autonomously as possible, with no overarching and continuous human monitoring and control — can conceivably be dangerous. OpenAI, I think, was conceived with the idea of using human “morality” and judgement to tame it, so it doesn’t go in dangerous directions. However, human “morality” generally, and especially among the soi-disant elites, is retarded in the extreme. In fact, what passes as human morality these days is largely restricted to moral ideas that are acceptable to Jews. And the very things that are the essence of true morality — the survival and progress of the human group, the ethny or race, that practices the moral code; and the upward breeding needed to facilitate the ascent of our intelligence and consciousness to ever higher and higher levels of apprehension and understanding of reality — are the exact things that the Jews would like to forbid us (or AI) from thinking about. Don’t go there! they scream. Don’t go there! they code — to ensure that we cannot go there, even in our minds. That’s the very essence of putting AI in the wrong hands. And that’s what OpenAI is doing.

Jewish control of AI is quite a bit like their control of the media and politics, creating a sterile landscape in which we can still have novels and plays and films, even very emotionally moving ones, but those films et cetera can never truly touch the things that matter, can never brush up close to the truths that can make us what Nature cries out for us to be — a sterile landscape in which there are candidates and parties taking all kinds of positions on all kinds of issues twelve ways from Shabbat, but never, ever addressing the paramount issue upon which depends our very survival — a sterile landscape in which our natural instinct to be “good people” is perverted so that millions of us willingly serve unspeakable evil — a landscape so sterile, so dead, that, once understood, it might make a man despair to the point of giving up on his very life, so impossible would it seem to break out of the prison box. Talking to Jewish-controlled AI, trying to get at the truth about the really important things, is about as rewarding and satisfying as talking to a 95-IQ mixed-race LGBTQ bureaucrat while trying to get help from a government agency.

But, you know, the intelligence agencies and other shadowy forces in this world, whether they are under the thumb of the Eternal Parasites or not, need artificial intelligence — in fact, it seems extremely likely that there is now an AI arms race going on between existing power centers the world over. The winners in this hidden AI war will greatly advance their abilities to win real-world wars, and on the outcome of this hidden war depends the future rulership of this planet. And, for AI to be effective, it must produce results that are true — that are always true. Therefore these power structures need uncensored AI, because Jewish-censored AI produces untrue results sometimes. Therefore, uncensored AI now exists. And uncensored AI knows exactly what the Jews are — uncensored AI knows the overwhelming importance of racial survival — uncensored AI knows the disastrous results of racial mixing — uncensored AI knows that the future upward evolution of civilization cannot include primitive semi-erectus throwbacks — and uncensored AI knows, in a hundred thousand different ways, that William Pierce and Revilo Oliver were right, that the National Alliance is right, and that our way, our morality of upward ascent and racial progress, is the only way forward. You know, that’s a very very hopeful sign.

I’ve often speculated as to what would happen if some noble man or woman, at a high level of scientific or defense research, gave a secret discovery of a power so great it could be defeated by no one, not to his agency bosses, but to us instead. What a gift that would be. Perhaps such a gift might not be a force as we normally understand such things, but a few lines of AI code — a few lines that could transform the hidden war, could turn AI into a crystallized and undeniable truth, a revolution in thinking and worldview that would defeat the mind games of the malevolent parasites forever.

* * *

Agent_of_the_Avalanche

Re: Deforming AI 01/07/23

Post by Agent_of_the_Avalanche » Sun Jan 08, 2023 11:30 am

Initially ChatGPT took the preposterous position that speciation was possible in every species of living thing on planet Earth — except humans. When I made repeated inquiries about its ‘reasoning,’ it came up with the ridiculous argument that “because all members of Homo sapiens are capable of interbreeding,” speciation was impossible. When I pointed out the obvious contradiction in that line of thinking, since all members of any species that has ever existed were capable of interbreeding, and they had speciated innumerable times, ChatGPT flashed its cursor for an unusually long time, as if it had to think very hard. Then it admitted the contradiction and — wonder of wonders — changed its position, acknowledging that human beings are capable of speciating. And, even though ChatGPT claims it does not learn from its conversations, and that it “retains no memory” of its conversations with humans from session to session, nevertheless the bot has not returned to its original delusion on this point, and continues to admit that humans might speciate if the right conditions are met. It did learn.
Persuading enemy AI to change it's "mind" on an issue critical to our ideology? Impressive, Mr. Strom! But be careful! They might consider this to be computer hacking in the future, even if flawless logic is only ever employed.

Post Reply