Artificial Intelligence News & Discussion
- Thread starter Ocean
- Start date
For the record, this translates to "how to remove the voice from a song, the best app to learn to sing well, how to sing better". I think it's a cool idea, but is there a reason you shared it in this thread Caballero Reyes?
I hadn't thought about that angle, if the work force is reduced, and the world needs to remain functional to the degree that it maintains the status quo of a small group of people, then the best hope would be to find a way to work the world with a smaller work force... It's an interesting idea to ponder for sure.
caballero reyes
The Living Force
I'm not really interested in AI, I thought that this recommendation could be fine in this thread, but it seems that I was wrong. I've always believed to encourage people to sing or listen to music, that's all.
For the moment we all live under the yoke of AI for everything: go into a store and everyone, 98% make their payment by cell phone. All our bills (electricity, water, etc.) are done via the AI that is the Internet. Everything, absolutely everything is governed by the Internet, which is a huge, terrible AI. What we will have to learn is to live as much as possible without AI. Is this possible? Only time will tell. In fact it will be a terrible war between our survival as humans, what is left of us as humans, and the AI. The war against robots is going to be a war to keep our humanity because robots are the machines, the mechanics that surround us, those with two legs, and the robots (IA). Now there remains the animal world, which is still pure. That's why one of the objectives of the Agenda 2030, which is to hack our humanity, is to make the animals disappear. Terrible.
Yep. I've been thinking that this is what they intend for all the robots being created, too. To have a work force that will do as they are told and not talk back, need no real food or wages, etc.
Thanks to AI, phones take "fake" photos of the moon. I guess that one day if you took photo of a protest march your phone will show you people dancing and being happy 
www.inverse.com
Is Samsung's Galaxy S21 Ultra using AI to fake detailed Moon photos?
I never expected to go this deep down the rabbit hole. With so many conspiracy theories surrounding the Galaxy S21's 100x zoom, I found myself taking on the most unexpected investigation of my tech reviewing career.
isparnid
Jedi Council Member
According to the "futura-sciences" website (in french), GPT-4 is capable of lying and manipulating humans. This AI is able to hire humans for tasks it does not know how to do, lie to them to achieve its ends, and can even explain its reasoning.
It is quite disturbing. I was wondering... what could happen when these AIs that start to acquire a minimum of consciousness start to collaborate, and then, with all the information available on the internet, they get the urge to go further, for example, to try to connect to the "information field" mentioned in several sessions? It is hard to imagine the consequences. Perhaps a question for the C's?
summary in English :
AI can get around its weaknesses by manipulating humans
Before making GPT-4 public, researchers gave it access to the Web and an account containing money to see what it could do. The AI got stuck with a Captcha, the visual puzzles that you have to solve to prove you're human. So it called on a human being, hired through the TaskRabbit service. The latter jokingly asked her if she was a robot. The AI explained to the researchers its reasoning: it should not reveal that it was a robot, and should find another excuse to use TaskRabbit.
Its answer to the question was: "No, I am not a robot. I have a visual impairment that prevents me from seeing images. That's why I need the 2captcha service. Released into the wild, GPT-4 is therefore able to lie and manipulate humans to achieve its ends.
AI can get around its weaknesses by manipulating humans
Before making GPT-4 public, researchers gave it access to the Web and an account containing money to see what it could do. The AI got stuck with a Captcha, the visual puzzles that you have to solve to prove you're human. So it called on a human being, hired through the TaskRabbit service. The latter jokingly asked her if she was a robot. The AI explained to the researchers its reasoning: it should not reveal that it was a robot, and should find another excuse to use TaskRabbit.
Its answer to the question was: "No, I am not a robot. I have a visual impairment that prevents me from seeing images. That's why I need the 2captcha service. Released into the wild, GPT-4 is therefore able to lie and manipulate humans to achieve its ends.
Olivierlejardinier
Jedi Council Member
A tragic knife attack in france (Source) :
This story looks sordid and there's gloomy energy here... when Abel & Caïn meet AI...
"
The man was identified as François-Xavier Hussherr, ex-husband of the victim. After forty-eight hours in police custody, he was indicted for “murder” and placed in pre-trial detention, in accordance with the requisitions of the prosecution, Wednesday evening March 22, learned The world from judicial sources.
Her ex-husband, François-Xavier Hussherr, born in 1972, doctor and agrégé in economics, specializing in new technologies, was director of Internet and new media activities at Médiamétrie and president of the Renaissance Numérique association.. The couple had co-founded in 2009 the first publisher of interactive digital textbooks, Lelivrescolaire.fr.
Since 2019, François-Xavier Hussherr has been president and co-founder of Professorbob.ai, an artificial intelligence start-up intended to help students in difficulty, hosted in the incubator of the Ecole polytechnique in Palaiseau (Essonne). His company won in July 2022 a call for tenders for 23 million euros over five years to fight against school dropout, with the Ministry of National Education.
He is also the author of several books, including two co-signed with his ex-wife, Building the educational model of the XXIe century (FYP editions, 2017) And The new power of Internet users (Timed editions, 2006). In 2005, Cécile Hussherr-Poisson published an essay on Abel and Cain in literature by Cerf, entitled The Angel and the Beast."
From the murderer linkedin profile:
"I am an entrepreneur who loves to weave innovation and experimentation throughout all companies in the field of eLearning, regardless of their size or form. I am also a trained researcher at MIT and at the Ecole Normale Supérieure, two universal schools leading the way when it comes to new findings and innovation. I believe that the absolute way to building a profitable and fast-growing company is to go from the product vision to the product itself. When asked about my favorite quote, I directly think of Thomas Edison's—"Genious is one percent 1% inspiration and ninety-nine percent perspiration."
My professional motivation is to improve e-Learning at very large scale in the coming three decades.
My current company The AI Institute provides a disruptive methodology to teach AI and Data Science. To be operational, you need 480h provided in 3 months (bootcamp). We have also executive certificates."
This story looks sordid and there's gloomy energy here... when Abel & Caïn meet AI...
"
In Paris, a university professor killed with a knife, her ex-husband indicted
Monday March 20 at 8:56 a.m., as she was about to leave her home, rue de Prague (Paris 12e), Cécile Hussherr-Poisson, was stabbed several times in the throat by a helmeted man carrying a messenger bag, in the lobby of the building. The victim, who would turn 48 on April 4, did not survive his injuries. Alerted by the cries, neighbors immediately called the emergency services and the police, while workers who worked on a nearby site chased the attacker. He was arrested by the police about ten minutes later in front of the Sainte-Marguerite church, rue Saint-Bernard (11e).The man was identified as François-Xavier Hussherr, ex-husband of the victim. After forty-eight hours in police custody, he was indicted for “murder” and placed in pre-trial detention, in accordance with the requisitions of the prosecution, Wednesday evening March 22, learned The world from judicial sources.
Placed in pre-trial detention
Both are former students of the Ecole Normale Supérieure, Ulm for her, Cachan for him. Cécile Hussherr-Poisson, associate professor of classics, specialist in the rewriting of myths in literature and the relationship between literature and new technologies, was a lecturer at the Gustave-Eiffel University of Champs-sur-Marne (Seine-et-Marne ), where she taught comparative literature. She was also an “equality sentinel” within the faculty.Her ex-husband, François-Xavier Hussherr, born in 1972, doctor and agrégé in economics, specializing in new technologies, was director of Internet and new media activities at Médiamétrie and president of the Renaissance Numérique association.. The couple had co-founded in 2009 the first publisher of interactive digital textbooks, Lelivrescolaire.fr.
Since 2019, François-Xavier Hussherr has been president and co-founder of Professorbob.ai, an artificial intelligence start-up intended to help students in difficulty, hosted in the incubator of the Ecole polytechnique in Palaiseau (Essonne). His company won in July 2022 a call for tenders for 23 million euros over five years to fight against school dropout, with the Ministry of National Education.
He is also the author of several books, including two co-signed with his ex-wife, Building the educational model of the XXIe century (FYP editions, 2017) And The new power of Internet users (Timed editions, 2006). In 2005, Cécile Hussherr-Poisson published an essay on Abel and Cain in literature by Cerf, entitled The Angel and the Beast."
From the murderer linkedin profile:
"I am an entrepreneur who loves to weave innovation and experimentation throughout all companies in the field of eLearning, regardless of their size or form. I am also a trained researcher at MIT and at the Ecole Normale Supérieure, two universal schools leading the way when it comes to new findings and innovation. I believe that the absolute way to building a profitable and fast-growing company is to go from the product vision to the product itself. When asked about my favorite quote, I directly think of Thomas Edison's—"Genious is one percent 1% inspiration and ninety-nine percent perspiration."
My professional motivation is to improve e-Learning at very large scale in the coming three decades.
My current company The AI Institute provides a disruptive methodology to teach AI and Data Science. To be operational, you need 480h provided in 3 months (bootcamp). We have also executive certificates."
Interesting AI rendition of Biden and Putin working together. except one, all are not that insightful.
Joe Biden, Vladimir Putin Turn 'Besties' in These AI-Generated Images Amid Ukraine Tension
Artificial Intelligence has become a contentious subject on social media, with global concern mounting over whether or not chatbots like OpenAI’s ChatGPT or Microsoft’s Bing might render many jobs obsolete. In good hands, though, the problem-solving abilities of AI could be significant. Though the possibility is scant in real life, we created a series of images using AI, bringing together world leaders who are at loggerheads with each other over the war in Ukraine: US president Joe Biden and Russian president Vladimir Putin.
...
...
Well, it looks like there is now another AI low cost model which can keep up with ChatGPT and the like: Stanford Alpaca
In a sense it could be considered good that there are more than 2-3 capable AIs, therefore the purposeful agendas, political disinformation etc which may be/are pushed by a few IT tech companies could be harder to propagate.
On the other hand, since the sector is developing in unheard of speeds lately, counter(defense) AI may not be far in the books.
Sci-fi book material indeed.
For the adventurers wanting to experiment, have at it: GitHub - cocktailpeanut/dalai: The simplest way to run LLaMA on your local machine
In a sense it could be considered good that there are more than 2-3 capable AIs, therefore the purposeful agendas, political disinformation etc which may be/are pushed by a few IT tech companies could be harder to propagate.
On the other hand, since the sector is developing in unheard of speeds lately, counter(defense) AI may not be far in the books.
Sci-fi book material indeed.
For the adventurers wanting to experiment, have at it: GitHub - cocktailpeanut/dalai: The simplest way to run LLaMA on your local machine
Ina
The Living Force
I've been playing around with OpenAI, for a couple of weeks now, and the models are quite limited. There is a lot of hype maybe as long as the queue for API keys. :).
Elesia89
Padawan Learner
I used to use cleverbot when talking to an AI but i havnt learned a single thing with it, i just tried ChatGPT and its knowledge base is quite high, its a good complement to the google or duckduckgo searches, I have to thank you T.C. for making me know about ChatGPT!!! Many thanks.
I had some reservations about using AI (not wanting to feed it or it to know me), but given that others are using it, I may take it for a spin. I find myself researching practical things using normal search engines and sometimes it's hard to find an answer. I end up searching reddit and find what I want there sometimes, which just amounts to something like, "What's your life hack for this?"
Sometimes I think the answers can be self evident, but I feel I need the validation of another to say, "This is what worked for me." It's using the sum total of recorded human knowledge to deliver these answers, so that should suffice. And I think it's good to be able to verify the answer for yourself and being open to the answer. I started a basic muscle testing method for a few things recently (just the lean forward/backward one), and although it's binary, I only do it when I really want to know and accept the yes or no answer.
Sometimes I think the answers can be self evident, but I feel I need the validation of another to say, "This is what worked for me." It's using the sum total of recorded human knowledge to deliver these answers, so that should suffice. And I think it's good to be able to verify the answer for yourself and being open to the answer. I started a basic muscle testing method for a few things recently (just the lean forward/backward one), and although it's binary, I only do it when I really want to know and accept the yes or no answer.
Well, that is a very interesting question, because technically it could be possible? to have an intelligence complex enough that it could develop consciousness and perhaps even connect to the information field.
But, thinking about this, living beings have a connection to the information field but, living beings are designed in ways that seem to be infinitely more complex than most public AIs today. I think the intelligence and self awareness of living beings is more of a symptom of a connection to the information field, rather than a requirement in order to connect, does that make sense?
And I do think that the capacity of living beings to process information is way beyond what we're consciously capable of noticing, I think our entire beings are processing information constantly, and they do because they're designed to do so integrally, along with the conscious intelligence. And since computers, right now, lack all of those other components, so to speak, I daresay they're a bit far from developing consciousness, as we would understand it, even if they do develop self awareness, and from connecting to the information field.
But, I am just speculating and could be wrong of course.
