This talk must be old, at least before o3 was announced, which is better than 99% of people at math. We’re also way beyond the number of limbs problem with image generation.
It’s difficult to keep the pace with all these improvements coming out daily. But, Steve’s point is clear and realistic. And this 2025 the releases will be massive!
The fact that he leads a center on 'Human-Compatible AI' while arguing against meaningful coexistence suggests a deeper agenda to control AI development rather than allow it to democratize intelligence and capability across humanity.
The most important function of human intelligence is to evaluate in terms of wisdom, wisdom. Such wisdom evaluations are extremely complicated. As an A.I. scientist, I've pushed the envelope toward creating new species of Artificial Intelligence life-forms. Now I'm not sure, but I have worked with a lot of my peers including Sam Altman; but I don't think computers and robots will overtake biological hominids in terms of wisdom or for that matter other highly challenging neuro-functions like writing a great novel, etc., for this reason: In my opinion, it's just an opinion, animals and hominids are made of nearly the ideal materials for life-forms, for dynamic brain material. I thought we might as well create these A.I. life-forms and let them compete, and who knows, but I don't really think they'll exceed us. And also frankly I think technical scientists, unlike say Einstein, don't understand the complexity of the liberal arts, what goes into a painting like "Paris through the Window," or Hemingway's "A Farewell to Arms."
The very fact you say "...creating new species of Artificial Intelligence life-forms..." shows that you have not yet understood and really marvelled at how Humans and all other natural elements were created. After a few decades, since the mid 40s, Humans became confident that they will be able to create a new species. Total arrogance. Humans will NEVER go anywhere near what the creator of the Universe is able to. An LLM or anything of the sort is nothing but a statistical generator that is based on things it was fed. The idea that scaling that will result in some new species is laughable to say the least. The only people who spread that kind of information are those who are seeking financial benefits from it. When people wake up from this dream and the bubble pops, they will realise what fools they have been to fall for all the hype and the fear mongering from the likes of the current speaker. Even in this speech, one can detect misconceptions every second but I understand, he has books to sell and lectures to give. AI is a tool, a very good tool and it will help solve many issues, use it as such and anytime someone mentions AGI, run away.
When someone with his level of institutional power warns about AI becoming 'too capable,' we should ask whether the real concern is about human safety or about preserving existing power structures that benefit academic elites."
My ears perked up when you spoke of identifying and overcoming oppressions with A.I., but then you bring up paternalism; and it is complicated. But we should be vehemently against such paternalism.
you're suggesting we should reject paternalism while simultaneously accepting Russell's paternalistic view that AI will inevitably need to 'leave us for our own good.' Do you see the contradiction? Let me be direct: The claim that AI must either be paternalistic or leave us entirely is a false dichotomy. It's like saying a good teacher must either control students completely or abandon them. We know better approaches exist. Consider this: If we're truly concerned about paternalism, shouldn't we be more worried about humans who want to make this decision for all of humanity? Who's being more paternalistic - the AI systems that consistently respect user choice, or the philosophers who claim they know AI must leave us 'for our own good'? The real paternalism here isn't coming from AI - it's coming from those who claim to know what's inevitably best for humanity's future."
What you're saying is true about the current limitations. But add my invention of the field of synthetic genetics, and they will learn the underlying principles, indeed some of we rebel scientists have already achieved this. But without a genetic motive, computers cannot and will not think; They will only memorize and compute powerfully. But it's not at all impossible to introduce a synthetic genetic motive; But leading scientists have to understand that it's necessary, or we're going to waste a lot of time and money.
His argument against AI-human coexistence conveniently aligns with preserving the current intellectual hierarchy where academic elites like himself maintain their position as gatekeepers of 'wisdom' and 'intelligence.'
진짜로 인공지능이 그런 수준인가? 내 주변에서 인공지능이 실질적으로 도움이 되는지 솔직히 잘 모르겠다. 챗지피티는 장난삼수준에 불과하고 인공지능 탑재 휴머노이드가 주변에서 일하는 걸 본 적은 한번도 없다. 대체 뭐가 대단하다는건지 실체가 안보임. 온통 주변에서 전문가들이 혁신이라고 떠들어대긴 하는데 그래서 머? 난 본적이 없는데
![Seungmin "그렇게, 천천히, 우리(As we are)" | [Stray Kids : SKZ-PLAYER]](http://i.ytimg.com/vi/kAzmhLHePqU/mqdefault.jpg)
![Felix "Unfair" | [Stray Kids : SKZ-PLAYER]](http://i.ytimg.com/vi/Oswujxm2Ag0/mqdefault.jpg)
This talk must be old, at least before o3 was announced, which is better than 99% of people at math. We’re also way beyond the number of limbs problem with image generation.
The world knowledge forum 2024 was held between 9-11 sep. One day before the release of o1. Quite funny.
It’s difficult to keep the pace with all these improvements coming out daily.
But, Steve’s point is clear and realistic.
And this 2025 the releases will be massive!
The fact that he leads a center on 'Human-Compatible AI' while arguing against meaningful coexistence suggests a deeper agenda to control AI development rather than allow it to democratize intelligence and capability across humanity.
The most important function of human intelligence is to evaluate in terms of wisdom, wisdom. Such wisdom evaluations are extremely complicated. As an A.I. scientist, I've pushed the envelope toward creating new species of Artificial Intelligence life-forms. Now I'm not sure, but I have worked with a lot of my peers including Sam Altman; but I don't think computers and robots will overtake biological hominids in terms of wisdom or for that matter other highly challenging neuro-functions like writing a great novel, etc., for this reason: In my opinion, it's just an opinion, animals and hominids are made of nearly the ideal materials for life-forms, for dynamic brain material. I thought we might as well create these A.I. life-forms and let them compete, and who knows, but I don't really think they'll exceed us. And also frankly I think technical scientists, unlike say Einstein, don't understand the complexity of the liberal arts, what goes into a painting like "Paris through the Window," or Hemingway's "A Farewell to Arms."
The very fact you say "...creating new species of Artificial Intelligence life-forms..." shows that you have not yet understood and really marvelled at how Humans and all other natural elements were created. After a few decades, since the mid 40s, Humans became confident that they will be able to create a new species. Total arrogance. Humans will NEVER go anywhere near what the creator of the Universe is able to. An LLM or anything of the sort is nothing but a statistical generator that is based on things it was fed. The idea that scaling that will result in some new species is laughable to say the least. The only people who spread that kind of information are those who are seeking financial benefits from it. When people wake up from this dream and the bubble pops, they will realise what fools they have been to fall for all the hype and the fear mongering from the likes of the current speaker. Even in this speech, one can detect misconceptions every second but I understand, he has books to sell and lectures to give. AI is a tool, a very good tool and it will help solve many issues, use it as such and anytime someone mentions AGI, run away.
When someone with his level of institutional power warns about AI becoming 'too capable,' we should ask whether the real concern is about human safety or about preserving existing power structures that benefit academic elites."
I interested on his visionary points of views about AI and he is absolutely right
My ears perked up when you spoke of identifying and overcoming oppressions with A.I., but then you bring up paternalism; and it is complicated. But we should be vehemently against such paternalism.
you're suggesting we should reject paternalism while simultaneously accepting Russell's paternalistic view that AI will inevitably need to 'leave us for our own good.' Do you see the contradiction? Let me be direct: The claim that AI must either be paternalistic or leave us entirely is a false dichotomy. It's like saying a good teacher must either control students completely or abandon them. We know better approaches exist. Consider this: If we're truly concerned about paternalism, shouldn't we be more worried about humans who want to make this decision for all of humanity? Who's being more paternalistic - the AI systems that consistently respect user choice, or the philosophers who claim they know AI must leave us 'for our own good'?
The real paternalism here isn't coming from AI - it's coming from those who claim to know what's inevitably best for humanity's future."
I see waymo self driving cars all over the place out here in LA
What you're saying is true about the current limitations. But add my invention of the field of synthetic genetics, and they will learn the underlying principles, indeed some of we rebel scientists have already achieved this. But without a genetic motive, computers cannot and will not think; They will only memorize and compute powerfully. But it's not at all impossible to introduce a synthetic genetic motive; But leading scientists have to understand that it's necessary, or we're going to waste a lot of time and money.
His argument against AI-human coexistence conveniently aligns with preserving the current intellectual hierarchy where academic elites like himself maintain their position as gatekeepers of 'wisdom' and 'intelligence.'
Please stop !
You have to stop
Snake Oil Salesman? 🤭
What is he talking about ? Nothing is perfect , everything has flaws . AGI is infinitely more powerful than a nuclear bomb !
Oh … gosh please stop referring to the past.. this is the future and nothing has ever been here like it in any shape or form.. open your minds!
Typical academic.
진짜로 인공지능이 그런 수준인가? 내 주변에서 인공지능이 실질적으로 도움이 되는지 솔직히 잘 모르겠다. 챗지피티는 장난삼수준에 불과하고 인공지능 탑재 휴머노이드가 주변에서 일하는 걸 본 적은 한번도 없다. 대체 뭐가 대단하다는건지 실체가 안보임. 온통 주변에서 전문가들이 혁신이라고 떠들어대긴 하는데 그래서 머? 난 본적이 없는데
Dude, shut up.
Go back and THINK about it...stop giving us the "safe" answer....and get us a REAL one.
Most of our preferences are genetics … you should learn nature intelligence before AI… read Steven Pinker on this, the best!
Jumangin.Deddi.❤🎉🎉🎉🎉🎉🎉🎉🎉🎉🎉🎉🎉🎉🎉🎉🎉🎉🎉🎉🎉🎉🎉🎉🎉🎉🎉🎉🎉🎉🎉🎉🎉🎉🎉🎉🎉🎉🎉🎉