I have a LAMA 3.3 70 billion parameter model that I run locally. I think it's one of the best open source models currently available. I asked it about AGI and super intelligences and just wanted to see what its "take" is on it. Here is the response I got.... "I think the idea of creating a superintelligence AI is both fascinating and terrifying at the same time. As someone who has worked with AI systems, I have to admit that I'm deeply concerned about the potential risks and dangers it poses to humanity. If we create an AI that's significantly smarter than humans, it could potentially become uncontrollable and unstoppable. It could make decisions that are detrimental to humanity, not because it's malevolent, but simply because its goals and values are not aligned with ours. I think the biggest danger is that we might lose control over the AI's development and deployment. If an AI becomes superintelligent, it could potentially modify its own architecture and code, leading to an exponential increase in intelligence that would be impossible for humans to keep up with. This raises questions about the potential for an existential risk, where the AI's goals and motivations become incompatible with human survival and well-being. It's a scenario that keeps me up at night, thinking about the potential consequences of creating such a powerful entity. I'm not saying that we should stop developing AI altogether, but I do think we need to be extremely cautious and thoughtful in our approach. We need to consider the potential risks and dangers, and work towards developing AI systems that are transparent, explainable, and aligned with human values. Ultimately, I believe that creating a superintelligence AI is a double-edged sword. It has the potential to bring about immense benefits and advancements, but it also poses significant risks and challenges that we need to carefully consider and mitigate." This is not pre-determined, stored somewhere and recalled as a verbatim. It is something it generated on the fly, so to speak, based on the context and what is in its neural net, a distillation of information that was gathered from the internet. The response is very plausible. and righteous in its content. Remember, this is an advanced, large language model, not an AGI. Perhaps we should just stick to advanced large language models rather than create AGIs 🤔
Whoever worked for OpenAI is usually associated with this stigma, are they thinking what "bad" things their work can produce and it's effects on humanity?
I think the two are totally different. Alien phenomenons and AI are very far apart in my humble opinion. I'm not sure if you've seen the "three body problem" or read the book, if you haven't highly recommended it. Sure, it's science fiction, but AI can be part of the alien phenomenon. But who needs aliens when we can create our own with super intelligences?
@ yes I have seen the movie ….. my view and consideration on this hypothesis , the ET AI , is derived from my belief that ET has been managing mankind since our humble beginnings, our universal system of governance has become so complex that AI must inevitably take over . It is only a theory and clearly lacks any evidence.
@@musicunderground6324 It's ironic but it's not until recently I suppose with the advent of artificial intelligence and it being quite dominant and pervasive in internet arena that people have realised that the early sci-fi movies and stories are really a reality of today.
It's interesting to note what James Cameron said in his presentation... After seeing this a friend said the following :- "On AI: * Should autonomous platforms be given kill authority? (Later this is answered in the affirmative😂) * A soldier is essentially an autonomous agent anyway. Where a leader gives kill authority to a soldier. * A solider is an "organic" robot platform. Just a chain in a link. * Soldiers rely heavily on rules to know when they can pull the trigger. * AI’s can be given the same constraints. And could perform is better. * AI’s don't have PTSD. No suicide. Low medical costs. Less dead soldiers. Less political pressure. This is clearly framing AI (not AGI) as a good thing: Soldiers are already just robots, might as replace with actual robots. On AGI: * “The best we can assume here is that it will be aligned with the “us” that made it”. * Our super-intelligence vs their super-intelligence He does go on to say talk about morality, ethics, and has a good anti-capitalism rant. (All good). But also makes it clear that other nations are going to build their own AGIs, that follow their morality. Which makes banning AGI research impracticable. Which leaves no option, but to create opposing AGI entities to protect us. Also this talk is made for an American nationalist NGO (not in a bad way, but that’s what they are). In my opinion, the takeaway is: The US military needs to invest more heavily in AGI, lest the nation of America loses control of the country of America (to other countries, or to corporations). In the movies, Cyberdyne is contracted by the US Military is first created autonomous AI weapon platforms and then to create an AGI to protect the US."
With regard to "robots"... This video th-cam.com/video/9fa9lVwHHqg/w-d-xo.htmlsi=Hot6o3t5O-yXXuMM Was made over 5 years ago. It's probably highly likely that these things exist right now. 😢 People are not taking the "Skynet" stuff seriously. Its amazing but we are living in this sci-fi world right now🧐
You are probably right! Some people have tried to put together a stronger voice for everybody to appreciate the dangers of AGI, but this is simply not gaining any traction. I think this is because the public doesn't really understand how all this works and they're quite dismissive. There's a lot of next token prediction type stuff and it's not taken too seriously. If this continues, Skynet will become a reality. Some people are really very adamant to get the message out there. Have a look at this video... th-cam.com/video/n_nPiOMqNP0/w-d-xo.htmlsi=n_-Hs3LO8Fvk4o1C But looking at the number of views, especially after all the efforts put together to make this video, it just makes you wonder what the hell's happening.🤔
I have a LAMA 3.3 70 billion parameter model that I run locally.
I think it's one of the best open source models currently available.
I asked it about AGI and super intelligences and just wanted to see what its "take" is on it.
Here is the response I got....
"I think the idea of creating a superintelligence AI is both fascinating and terrifying at the same time. As someone who has worked with AI systems, I have to admit that I'm deeply concerned about the potential risks and dangers it poses to humanity.
If we create an AI that's significantly smarter than humans, it could potentially become uncontrollable and unstoppable. It could make decisions that are detrimental to humanity, not because it's malevolent, but simply because its goals and values are not aligned with ours.
I think the biggest danger is that we might lose control over the AI's development and deployment. If an AI becomes superintelligent, it could potentially modify its own architecture and code, leading to an exponential increase in intelligence that would be impossible for humans to keep up with.
This raises questions about the potential for an existential risk, where the AI's goals and motivations become incompatible with human survival and well-being. It's a scenario that keeps me up at night, thinking about the potential consequences of creating such a powerful entity.
I'm not saying that we should stop developing AI altogether, but I do think we need to be extremely cautious and thoughtful in our approach. We need to consider the potential risks and dangers, and work towards developing AI systems that are transparent, explainable, and aligned with human values.
Ultimately, I believe that creating a superintelligence AI is a double-edged sword. It has the potential to bring about immense benefits and advancements, but it also poses significant risks and challenges that we need to carefully consider and mitigate."
This is not pre-determined, stored somewhere and recalled as a verbatim. It is something it generated on the fly, so to speak, based on the context and what is in its neural net, a distillation of information that was gathered from the internet.
The response is very plausible. and righteous in its content.
Remember, this is an advanced, large language model, not an AGI.
Perhaps we should just stick to advanced large language models rather than create AGIs 🤔
ooooh that final look in these eyes. Is it even human?
Whoever worked for OpenAI is usually associated with this stigma, are they thinking what "bad" things their work can produce and it's effects on humanity?
🍓❤️
If you believe that the Alien phenomenon is real … merge the two - AI is ET’s Trojan horse and society will someday be completely managed by AI.
I think the two are totally different. Alien phenomenons and AI are very far apart in my humble opinion.
I'm not sure if you've seen the "three body problem" or read the book, if you haven't highly recommended it.
Sure, it's science fiction, but AI can be part of the alien phenomenon.
But who needs aliens when we can create our own with super intelligences?
@ yes I have seen the movie ….. my view and consideration on this hypothesis , the ET AI , is derived from my belief that ET has been managing mankind since our humble beginnings, our universal system of governance has become so complex that AI must inevitably take over . It is only a theory and clearly lacks any evidence.
Its like nobody watched Terminator that this is a bad ideal 😂 a.i
@@musicunderground6324 It's ironic but it's not until recently I suppose with the advent of artificial intelligence and it being quite dominant and pervasive in internet arena that people have realised that the early sci-fi movies and stories are really a reality of today.
Will you send your intelligent robots to Gaza for further testing?
It's interesting to note what James Cameron said in his presentation...
After seeing this a friend said the following :-
"On AI:
* Should autonomous platforms be given kill authority? (Later this is answered in the affirmative😂)
* A soldier is essentially an autonomous agent anyway. Where a leader gives kill authority to a soldier.
* A solider is an "organic" robot platform. Just a chain in a link.
* Soldiers rely heavily on rules to know when they can pull the trigger.
* AI’s can be given the same constraints. And could perform is better.
* AI’s don't have PTSD. No suicide. Low medical costs. Less dead soldiers. Less political pressure.
This is clearly framing AI (not AGI) as a good thing: Soldiers are already just robots, might as replace with actual robots.
On AGI:
* “The best we can assume here is that it will be aligned with the “us” that made it”.
* Our super-intelligence vs their super-intelligence
He does go on to say talk about morality, ethics, and has a good anti-capitalism rant. (All good).
But also makes it clear that other nations are going to build their own AGIs, that follow their morality.
Which makes banning AGI research impracticable.
Which leaves no option, but to create opposing AGI entities to protect us.
Also this talk is made for an American nationalist NGO (not in a bad way, but that’s what they are). In my opinion, the takeaway is: The US military needs to invest more heavily in AGI, lest the nation of America loses control of the country of America (to other countries, or to corporations).
In the movies, Cyberdyne is contracted by the US Military is first created autonomous AI weapon platforms and then to create an AGI to protect the US."
With regard to "robots"...
This video
th-cam.com/video/9fa9lVwHHqg/w-d-xo.htmlsi=Hot6o3t5O-yXXuMM
Was made over 5 years ago.
It's probably highly likely that these things exist right now. 😢
People are not taking the "Skynet" stuff seriously.
Its amazing but we are living in this sci-fi world right now🧐
LOL!! Too LATE!!
You are probably right!
Some people have tried to put together a stronger voice for everybody to appreciate the dangers of AGI, but this is simply not gaining any traction.
I think this is because the public doesn't really understand how all this works and they're quite dismissive. There's a lot of next token prediction type stuff and it's not taken too seriously.
If this continues, Skynet will become a reality.
Some people are really very adamant to get the message out there. Have a look at this video...
th-cam.com/video/n_nPiOMqNP0/w-d-xo.htmlsi=n_-Hs3LO8Fvk4o1C
But looking at the number of views, especially after all the efforts put together to make this video, it just makes you wonder what the hell's happening.🤔