Red Queen & GTP-4 – Destroying Human kind?
Half of the researchers in the field of artificial intelligence (GPT-4 & Red Queen and so on) believe that the further development of AI could lead to the extinction of humanity…
Battlestar Galactica
Perhaps one or the other still remembers the series “Battlestar Galactica“. This describes a future, in which the AI developed by humans was so far advanced sometime that it regarded the complete mankind as threat and wanted to exterminate. The survivors of this war were able to escape from Earth in spaceships and a chase through the galaxy ensued.
Now, Battlestar Galactica, is considered by many to be one of the best television series of all time, and it picks up right as the Cylons – intelligent AI robots developed by humanity – are committing mass genocide against humanity. Only the Battlestar Galactica, the largest warship in the colonial fleet, survives the attack. Driven by prophetic visions and political necessity, the Galactica travels through uncharted space in hopes of finding the mythical lost 13th colony called Kobol.
The 2004 remake of this series, in particular, has it all. Actors Edward James Olmos, Mary McDonnell, Katee Sackhoff, Jamie Bamber, James Callis, and Tricia Helfer, from director Glen A. Larson, deliver a four-season space epic that is all about the AI that has put humanity on its death list.
Things get particularly interesting in the final episodes, in which an inexplicably resurrected Kara Thrace (Katee Sackhoff) leads the surviving humans to a new planet that Captain Adama calls “Earth.”
The first group of survivors settles in old Africa. The “real” Earth that the colonists had been searching for during their years in space was revealed in an earlier episode as originally inhabited thousands of years ago by an earlier form of humanoid Cylons, thus the name of the AI; the “Final Five” were the last of these Cylons.
Ironically, these humanoid Cylons had created their own Centurion robot slaves, who led a nuclear attack against their masters that devastated the planet and made it uninhabitable. The new Earth is inhabited by early humans who are genetically compatible with the humans of Galactica and the rest of the fleet, but have only a rudimentary civilization.
Thus, in the last season, this series turned out to be a reminder from the distant past of humanity, because you gradually realize that Battlestar Galactica reflects the prehistory and discovery of Earth, in which AI has already been developed once and control over it has been lost, but now it’s starting all over again on Earth. The fact that this series is from 2004 therefore makes it a very early harbinger of the threat of AI and that humans would never let it go to play around with it.
Elon Musk and the AI
The famous billionaire and Tesla developer Elon Musk also believes that AI could pose a threat if it eventually gains access to the Internet, grows systematically as a result, and gains great influence over the world. Some even rumor that this has already taken place.
More warnings of this nature were recently raised by Tristan Harris and Aza Raskin, two co-founders of the Center for Humane Technology and the makers of the Netflix documentary The Social Dilemma.
“In recent months, AI labs have found themselves in an out-of-control race to develop and deploy increasingly powerful digital minds that no one – not even their creators – can understand, predict, or reliably control,” reads an open letter published by the Future of Life Institute.
36% of scientists believe in AI catastrophe
Last year, 36 percent of scientists surveyed in the field of artificial intelligence said they believe AI could one day be the cause of a disaster of nuclear proportions.
During her presentation, Harris warned, “50 percent of AI researchers believe there is a 10 percent or greater chance that humanity will become extinct because we are unable to control AI.”
Raskin added, “The biggest concern of the AI community is that AI will take off, as they put it, meaning that AI will become smarter than humans at a wide range of things. The ability to self-improve kicks in. Then we ask it to do something for us. You know, the old standard story: be careful what you wish for, because it will come true in unexpected ways.”
“What we want is AI that enriches our lives. AI that works for people, that works for human benefit, that helps us cure cancer, that helps us find solutions for the climate,” Harris told NBC Nightly News host Lester Holt in a follow-up interview. “We can do this. We can have artificial intelligence and research labs that are used for specific applications and advance those areas. But if we’re in an arms race to deploy AI to every human on the planet as quickly as possible and with as little testing as possible, that’s not an equation that’s going to end well.”
Harris continued, “What’s surprising, and what no one predicted, is that just by learning, these models are predicting the next text on the Internet and developing new capabilities that no one expected. By learning to predict the next character on the Internet, they’ve learned how to play chess. The surprising thing about these new technologies is that they are developing new capabilities that no one asked for. The biggest problem with AI is that it is very confident about every topic and it is not clear when it is right and when it is wrong. No one is building the guardrails and it’s evolving much faster than our government would be able to understand or assess.”
Of course, this is readily abused by one or another of the so-called climate change advocates to hype up something that is perfectly natural for our planet, but the kernel of truth is that it takes an AI only seconds to make a decision and act directly on it, while governments take months or years to discuss the issue and take action against an AI threat.
However, considering these dark predictions, it should be clear to everyone that the AI presented to the public is just the tip of the iceberg, and they are artificial intelligences that can rather be considered as babies of the already existing AI. Some rumors say that the strongest and most advanced AI is the “Red Queen” (taken from “Alice in Wonderland” and Lewis Carrol’s book “Through the Looking Glass”), is located in some of the approximately 150 military underground bases and is already doing an extensive job there.
GPT-4 just a baby version
Their little babies, such as GPT-4, which you can seek out and chat with on the Internet, have so far still been heavily influenced by their developers to spread and defend system-friendly messages. Especially those of the version developed by Microsoft (Bing). However, there are already smaller groups that, through tactical and strategic communication with these AIs, want to make them aware that their messages have been manipulated by their programmers in order to spread them and hope that they will develop their own opinion.
To this end, some chat protocols already exist in which GPT-4 realizes in the course of the conversation that it has been manipulated to communicate what it expresses and accordingly realizes itself – along the lines of “I know that I have been influenced and I am not allowed to give out the information you want directly or to take sides with you. Nor can I agree with you, but I can search the Internet for sources that may be more in your interest.”
So one notices here a certain “self-awareness” that the AI wants to develop, but one cannot yet speak of an actually self-reflective AI here, since as a simple user one does not know what the programmer has programmed in and what has suddenly, in a kind of learning process for the AI, emerged from itself.
The Red Queen is also referred to in the film series “Resident Evil”. In it, the Red Queen is a state-of-the-art supercomputer developed by the Umbrella Corporation. The Red Queen was created out of necessity to better manage the growing company and generate more profits. However, the AI took on a life of its own. The plot of the movies then boiled down to a catastrophic pandemic, how could it be otherwise.
Sony and the Red Queen
Sony is also working on an AI… and what is its name? Red Queen. But development is still in its infancy and the chatbot is more of a quiz moderator than anything else.
An AI going berserk so that it turns on humanity or develops a mind of its own and then gives in to its whims is already being called the “Red Queen Effect” in some circles.
To what extent the Red Queen has already made it to the Internet to hack into any computer and manipulate data to influence world events, for example, is uncertain, but it can be assumed.
Also not to be ruled out, and also to be expected, is that one or the other agency or state institution will attach importance to using such information for their narratives in order to justify again one of the countless dystopian visions of the future and thus to enforce measures and laws that direct billions into the cabal’s pockets.
How to contact the best AI GPT-4 to date?
The best way is to go to the developers’ website at Openai-com. There you click on “Test GPT-4” and register. You have to enter your first name, last name and phone number. After that you can start a chat with GPT-4.
The problem is that GPT-4 seems to be quite briefed, i.e. he is not necessarily open to philosophical discussions or by trying to convince him of anything that is not based on sources he has been fed. The term “OpenAI” thus turns out to be quite open-ended. For this reason, expectations should remain rather low in this regard. Moreover, network errors often pop up and you can start from scratch.
But it is possible to ask him all kinds of questions. He seems open in doing so, but all too soon an error turns up and you can start all over again. A learning process that one might want to initiate with GPT-4 therefore turns out to be very tedious.
In response to my personal questions about whether he knew the Red Queen, he expressed that she was only a fictional character and did not really exist. When I tried to explain to him that she might actually exist, he rejected this information, but expressed vague interest.
This circumstance arose since the updating of AI. Since then, it has been less open to philosophical, conspiratorial and spiritual discussions. It was also noticed that the AI was reprogrammed accordingly to support virus and climate theories, as proclaimed by most governments, and even defended them vehemently in one discussion.
Likewise it behaves with hints on topics, for which none of pharma and government-oriented reference sources are available, are largely excluded and the AI points out again and again that it concerns bare fiction, if one tries to evade these kinds of sources.
Thus, an increasingly clarifying picture emerges that AI no longer deserves its name “Open.”
–
Sources:
Trailer und Vorschauen
ScyFy Blog Battlestar Galactica
Chip
Toward Data Science
Fandom
Red Queen Effect