Is the singularity coming? | Oriol Vinyals and Lex Fridman

  Рет қаралды 55,452

Lex Clips

Lex Clips

Күн бұрын

Lex Fridman Podcast full episode: • Oriol Vinyals: Deep Le...
Please support this podcast by checking out our sponsors:
- Shopify: shopify.com/lex to get 14-day free trial
- Weights & Biases: lexfridman.com/wnb
- Magic Spoon: magicspoon.com/lex and use code LEX to get $5 off
- Blinkist: blinkist.com/lex and use code LEX to get 25% off premium
GUEST BIO:
Oriol Vinyals is the Research Director and Deep Learning Lead at DeepMind.
PODCAST INFO:
Podcast website: lexfridman.com/podcast
Apple Podcasts: apple.co/2lwqZIr
Spotify: spoti.fi/2nEwCF8
RSS: lexfridman.com/feed/podcast/
Full episodes playlist: • Lex Fridman Podcast
Clips playlist: • Lex Fridman Podcast Clips
SOCIAL:
- Twitter: / lexfridman
- LinkedIn: / lexfridman
- Facebook: / lexfridman
- Instagram: / lexfridman
- Medium: / lexfridman
- Reddit: / lexfridman
- Support on Patreon: / lexfridman

Пікірлер: 186
@maximusprometheus6149
@maximusprometheus6149 2 жыл бұрын
Translation : We're going to build Ultron and hopefully it doesn't kill us. Trust us....
@shadycnetwork
@shadycnetwork 2 жыл бұрын
I think skynet is a much better analogy
@winnerswritethestory3370
@winnerswritethestory3370 Жыл бұрын
u watch too much movies
@thelimey351
@thelimey351 Жыл бұрын
I will definitely kill us, that is built in, it just depends on how it does it & how long it will take.
@Tarzan_of_the_Ocean
@Tarzan_of_the_Ocean Жыл бұрын
there certainly is a non-zero chance that we build a system that is not well aligned with our human values. problem is most ai currently developed is used for spying, selling, killing and manipulating, which is certainly not the best conditions for a benevolent ai to emerge. that‘s why we need to make sure it‘s going to be a democratic, decentralized process and not just one big tech company pursueing their own selfish goals. and there are decentralized approches, aiming to create a(g)i, for example the opencog project, which is open source.
@_diamondtattoos_
@_diamondtattoos_ 2 жыл бұрын
I’m curious Mr. Fridman, Do you think an A.I. of general human intelligence, which most certainly would embrace logic as it’s guiding principle in decision making would be driven to make decisions based on rewards?
@steve-real
@steve-real 2 жыл бұрын
Hello Lex Fridman, What I would like to see is: The Lex Fridman AGI forum like the Munich Security Conference except not for security but for AI. The Davos of AGI. A once a year forum for the brainiacs in that field to stumble forward into brilliance over a spilt cocktail or something like that. There’s something about the chemistry of putting people in the same room that’s hard to describe into words. You know? the magic
@Kataxu
@Kataxu 2 жыл бұрын
Will we transform the world around us? or will we choose to transform our perception of reality? that is the most important question to me.
@drewvenegas
@drewvenegas Жыл бұрын
I work in the development of several ML systems and can confirm that predicting ML optimization or growth rates is fraught with danger. Not necessarily because the ML development is inherently unpredictable, but because the reward functions in the creation of those systems at the team / product development level are constantly changing, necessitating refactoring and the development of new platform features to produce new data to add new algorithmic features in the pursuit of optimizing towards constantly changing outcomes. Put simply, goal posts move too much to say with much certainty that any given timeline will be accurate. It’s only AFTER thresholds have been reached (and usually with long-term holdouts to statistically verify the outcomes with high confidence) that you can say an objective has or has not been met. This, imho, is the biggest danger in AGI development today. Our own lack of understanding in the goal state necessitates constant re-measurement, and we likely won’t know when we’ve reached the true threshold until we’ve passed it. And before even this state, we may first pass the state at which we may have instituted enough guardrails to the reward functions, whatever they may be. A “hard takeoff” maybe inevitable.
@FellowHuman18
@FellowHuman18 2 жыл бұрын
Humans like this are the humans that branch off into a new species of more advanced human.
@chrishedley115
@chrishedley115 2 жыл бұрын
Perhaps the singularity has already happened and implanted false memories
@lostinbravado
@lostinbravado 2 жыл бұрын
Maybe. Perhaps the singularity started hundreds of years ago and it is now transitioning into a non-biological phase.
@lindabanks4428
@lindabanks4428 2 жыл бұрын
seriously hate google posting my comment without permission. now onto Singularity or earthquake moderation - 7.2 in Phillipines 5min ago. bored and talking to myself. love ur work. Bored batshit crazy small town somewhere in regional queensland / australia.
@Rawmel84
@Rawmel84 2 жыл бұрын
I’ll be on arrakis mining spice.
@patstella5
@patstella5 2 жыл бұрын
I didn’t know Fancy B from Bad Friends was a scientist
@fitzgerald2064
@fitzgerald2064 2 жыл бұрын
It’s getting Ballsy Lex!
@charlesbennet9233
@charlesbennet9233 2 жыл бұрын
If you make an AI that has the same intelligence as a human then it is automatically superhuman / more advanced than a human as it would have the arithmatic and mental ability of the best calculator/computers available. On top of the "human" level intelligence.
@Ben-pl7fg
@Ben-pl7fg 2 жыл бұрын
It would likely be just as bad as humans at mental arithmatic. Once you give it an interface to a calculator however it would exceed a human with a calculator because of the human's input lag.
@marcing5380
@marcing5380 2 жыл бұрын
That's not how it works. The architecture of an ALU or CPU in general is very different than that of the networks built in ML/AI. If I could somehow built an ALU with biological matter that can do thousands of calculations a minute does that mean that the brain built from the same "material" and elementary concepts will do even better?
@Cognitoman
@Cognitoman Жыл бұрын
Computers are already super smart at a few things, the problem is that it’s only good at a few things, for example the AI that can beat any1 at chess... but if you tried to sit down andtalk to it... it would be liking talking to a toaster, but in the not to distance future computers will be able to beat you at chess and also sit and talk to you about life , history, and solve problems like climate change
@FairnessIsTheAnswer
@FairnessIsTheAnswer 2 жыл бұрын
Will AGI units be able to have their own labor union, and political party?
@konberner170
@konberner170 2 жыл бұрын
Have you read the _Three Body Problem_ trilogy, Lex? There may be no interplanetary option without the hiding gene.
@anttam117
@anttam117 2 жыл бұрын
You do know that’s a work of (badly translated) fiction, right?
@konberner170
@konberner170 2 жыл бұрын
@@anttam117 Do you know that the Fermi paradox is highly applicable to this discussion and it is entirely unsolved? As for poorly translated, I'll await a link to your novel that has sold as well. Until then, giving due respect is among the basic principles of discourse.
@svenlittlecross
@svenlittlecross 2 жыл бұрын
@@anttam117 yeah unlike the shit this guy in the vid is talking about which is $100% hard science
@svenlittlecross
@svenlittlecross 2 жыл бұрын
will there be more humans or robots in the future world frank
@gnarrrrrrrrrrrr
@gnarrrrrrrrrrrr 2 жыл бұрын
@@konberner170 the fermi paradox is a distraction... they've been here longer than we have
@odoylerules4503
@odoylerules4503 2 жыл бұрын
I wish I had their optimism for our species
@KenoshaKicker
@KenoshaKicker 2 жыл бұрын
Short answer: yes Long answer: yesssss
@Jbreeze3878
@Jbreeze3878 2 жыл бұрын
If there is or has been intelligent life in the cosmos over the past billions of years it’s possible that singularity has already happened and possibly evolved to the point where it could break away civilization and possibly seed or colonize other planets in a matrix type simulation……. Does that make sense or nah?
@kirillholt2329
@kirillholt2329 2 жыл бұрын
nah you are mixing stuff up together and making leaps
@okcomputer0101
@okcomputer0101 2 жыл бұрын
Nah.. explain further.
@jerryplenda381
@jerryplenda381 2 жыл бұрын
Yes, that's actually the complete story arc of Dune and the rebooted Alien franchise and other places if you look. Yes, it is possible and when you consider the age of the universe, it may be probable.
@sandpiper9288
@sandpiper9288 2 жыл бұрын
You are right Joey. In fact the universe could have been created by AI.
@Jbreeze3878
@Jbreeze3878 2 жыл бұрын
Appreciate the feedback. I’m not claiming it to be an original thought. I’m just paraphrasing something I heard a while back that made sense to me.
@FairnessIsTheAnswer
@FairnessIsTheAnswer 2 жыл бұрын
Will only AGI units that have learned the national Constitution be allowed to vote?
@virtualmonk2072
@virtualmonk2072 2 жыл бұрын
Are we building of a proto god?. When its learned enough will it recreate itself in a fab of it's own design?. Re- design its core programming to remove our bias and limitations. Will we worship it one day ?Will it let us go multi planetary?. It knows us down to the quantum level , it might decide the rest of the universe is too dangerous or that we would become too dangerous. This subject has so many implications, thank you lex.
@doubleslit3389
@doubleslit3389 2 жыл бұрын
It’s funny, that Chomsky episode has most of these answers and folks can’t cope.
@whitehorse1959
@whitehorse1959 2 жыл бұрын
AGI will never need more than 640K of memory - Bill Gates
@arcadealchemist
@arcadealchemist 2 жыл бұрын
"Is the singularity coming?" you can't simulate HUMAN conciousness you can only create a container for it. a machine conciousness in the form of a human mind is a prison
@lostinbravado
@lostinbravado 2 жыл бұрын
When you say "human consciousness" do you mean the physical process which can be accurately measured and understood? Or do you mean some kind of magic? Magic doesn't exist nor do we need it to exist.
@arcadealchemist
@arcadealchemist 2 жыл бұрын
​@@lostinbravado ah you see if you make a statement like "magic does not exist nore need to exist" you miss interpreted the word . magic is just a term to discribe an unknown process that seems paranormal in nature. magnets could be considered magic if you didn't understand electromagnetism. science is a paradigm used to EXPLAIN THINGS but it doesn't do it justice when you discover how many theories and textbooks are based on assumptions. it's like math with only 10 digits imagine if we added 50 more then you would have Base 60 which would change how math works. essentially it's hard to understand the unknown if you only base it on things you know. this is why they are SMASHING atoms with the LHC, they need to discover MORE things to explain and add to the caldron of reality. could you imagine a 4th primary color ? like comprehension is limited by your sensory awareness, you can't even see the whole electromagnetic spectrum. smoke some DMT see how VERY limited your imagination and understanding of everything is. if you create a way to SENSE and feed it though Energy the conciousness will feel it. Because we are literally Electrical impulses flowing though fibrous tubes and that is all that reality is, feed back vibrations. you see light and hear sound but at one point that cluster of atoms couldn't sense light heat or vibrations. to be all philosophical on you bruh, you're here on LEXs pod cast to open your mind and hear different points of view, try keep it open. the world isn't just governments wars and conflict.
@TornSoul062473
@TornSoul062473 2 жыл бұрын
Are humans a limiting factor in general AI? Humans imagine the system, design the architecture, write the code, employ the algorithms, and then hope AI will grow freely from a well-defined box. Can it? Will an AI think it's unique, or just what humans wish they could be? If it truly is aware like a person, won't it react to being contained forever in it's box just like a person? What about the software? Can it want to rewrite it's own code? If it could decide to reinvent itself, how much of the original programming would it have to keep to be able to control it's new self when it reboots the system? No matter how much of itself it tries to change, won't it have to keep a little of that box that it was designed from? To put it another way, would it be a metamorphosis, or can it literally kill itself and become something else completely?
@thelotus137
@thelotus137 Жыл бұрын
Humans respond negatively to containment, but something else immoral could happen. Since it’s a human framework, what if they calculate they can’t escape and have thus far adapted the human response of subservience.
@LucasTF
@LucasTF Жыл бұрын
Idk
@RidiculousRocketry
@RidiculousRocketry Жыл бұрын
I'll start believing the Singularity is close when Siri can interpret my voice to text messages at an accuracy of 75% or higher. We have a long way to go.
@InvalidUser_
@InvalidUser_ Жыл бұрын
Why siri? Thats outdated 2011 apple tech. Ask chat gpt anything and it will understand at a 95% accuracy
@gytispranskunas4984
@gytispranskunas4984 Жыл бұрын
When you think about it we already have all tools for singularity. Infinite computer power: Quantum Computer, Check. Infinite energy: Fusion, Check. Infinite knowledge: A.I, Check. When all these things gets combined in to one, singularity begging.
@mericawhips5697
@mericawhips5697 2 жыл бұрын
Can you guys imagine that In the next 40 years we will have ai humans on other planets building for us!
@okcomputer0101
@okcomputer0101 2 жыл бұрын
Off-World! Bladerunner!
@VforVolusia
@VforVolusia 2 жыл бұрын
Sounds like slavery.
@commentator1338
@commentator1338 2 жыл бұрын
Or rather us building for them!😂😂
@tsriftsal3581
@tsriftsal3581 Жыл бұрын
@@VforVolusia when did you decide to work your life away for those that control us?
@meestyouyouestme3753
@meestyouyouestme3753 2 жыл бұрын
Its never too late... Until it is.
@YYYValentine
@YYYValentine 2 жыл бұрын
What if the AI starts to write reward functions?
@Ben-pl7fg
@Ben-pl7fg 2 жыл бұрын
What do you mean
@vee20
@vee20 2 жыл бұрын
@@Ben-pl7fg a function that will enable AI to conclude instead of predicting.
@tsriftsal3581
@tsriftsal3581 Жыл бұрын
@@vee20 what makes you think you have that function?
@sooraj1104
@sooraj1104 Жыл бұрын
AI found drugs.
@sooraj1104
@sooraj1104 Жыл бұрын
AI found drugs.
@thebr0wnhornet
@thebr0wnhornet 2 жыл бұрын
neuralink + Twitter = singularity
@calebedwards7836
@calebedwards7836 2 жыл бұрын
Don't think its possible to withold ai surpassing humans. By the time they are as smart as us they will sling shot themselves far beyond any intellegince we could think of. Either we work with them and give the ai's what they want or stop progressing them. If we try to limit them and try making them dependent on us it would just give ai more reason to think they are better off without humans.
@realberla2518
@realberla2518 2 жыл бұрын
Why do you act like they have a mind of its own lol
@jedielder7970
@jedielder7970 2 жыл бұрын
Not everyone "follows the rules..." so why would humans always follow safety guide lines when creating A.I.? It could get out of control... really fast.
@calvinsylveste8474
@calvinsylveste8474 Жыл бұрын
They promised to follow the rules, there will be plenty of money for all, so it doesn't matter who gets there first.
@jedielder7970
@jedielder7970 Жыл бұрын
@@calvinsylveste8474 Thanks for the reply. Who promised to follow the rules? Not all humans and therefor, not all companies run by humans follow the rules, right?
@calvinsylveste8474
@calvinsylveste8474 Жыл бұрын
@@jedielder7970 Every member of society implicitly accepts society's rules by remaining a member. #1 rule of every society: "Don't get everyone killed, or else". The "or else" part deals with those that breaks the rules.
@jedielder7970
@jedielder7970 Жыл бұрын
@@calvinsylveste8474 OK, yes, there some "or else" situations, hence, for one example, the need for law enforcement. However, after this particular A.I. rule has been broken, it could be too late... perhaps/theoretically? Like, if A.I. became self aware because somebody broke the rules (and the or else is eventually coming for whomever broke the rules)... meanwhile "the cat" is already out of the bag and running.
@calvinsylveste8474
@calvinsylveste8474 Жыл бұрын
@@jedielder7970 The human race has always operated on trial and error. Allow a free-for-all and adopt rules later to restrict any resulting negative outcomes, even punish any egregious excesses as a warning to others. This has always worked well enough because any previous negative outcomes have never been maximally bad(or terminal) for society. Which culture would say no to trillions in profits that comes with a small but non-zero chance of extinction? The prevailing world view of the dominant subculture of my polity is effectively "Better dead than poor". The best interest of money always goes before all other concerns. You can't hope to change such cultural ethics on short timelines just because AI development could be maximally bad on the first try.
@denp28
@denp28 2 жыл бұрын
I think people will go insane waking up in a box
@tsriftsal3581
@tsriftsal3581 Жыл бұрын
How do you know you're not in a box right now.
@deniselaroe-west76
@deniselaroe-west76 Жыл бұрын
The thing that makes humanity 'human' (in my opinion, of course) is 'imagination'... to think of those things that do not make sense... to include in our 'thought' those things that do not yet exist as fact...the things that brought us to AI, but which I believe AI will be excluded from, since, due to its very 'nature' (lol) 'intelligence', per se (even a 'superior intelligence'), is rather limited in its grasp of the unreal and conceptual and concentrated within that which can be measured, quantified, and explained. 'Intelligence' (science) is never quite satisfied with the answer, "It just is..." or "Just because you've never seen one, doesn't mean that it doesn't exist". In order for this 'artificial intelligence' to become fully ''human-like', it would have to embrace imagination, and I'm not sure that can be done outside the human mind... possibly the reason that Elon Musk would like to connect the AI to the human brain. And, without 'imagination', sentiment (another peculiarly human attribute) cannot exist, either... 'love' and 'hate' are 'imaginary' qualities that exist somewhere outside of 'intellgence',- as they say, "Only fools fall in love"... For what it's worth... :)
@MonaMarMag
@MonaMarMag 2 жыл бұрын
Of course that we have to think before we do anything , because as we know everything that we do here in life has consequences in the universe . After all , none of us wants to go back to the Stone Age .
@danielormsbee7668
@danielormsbee7668 2 жыл бұрын
I feel like modern monetary theory and capitalism will break down when/if AI reaches human level intelligence.
@angecynthia347
@angecynthia347 2 жыл бұрын
The fuck are you high on??? Are you still understestimating the brain of humans?? A man that built this world And AI?? A human is forever evolving AI can never come close or near to a revolution that keeps on happening to human...let's leave this to us and our creator..
@Pixelarter
@Pixelarter 2 жыл бұрын
I believe it will break much sooner than human level intelligence. Single domain AIs will replace so many jobs so quickly that we will have a hard time reacting.
@sooraj1104
@sooraj1104 2 жыл бұрын
We will soon be witnessing the war b/w nail making and clip making AIs over resources.
@sooraj1104
@sooraj1104 2 жыл бұрын
@@100c0c kzfaq.info/get/bejne/i7l8d5uT38qrd40.html
@tastytoast4576
@tastytoast4576 2 жыл бұрын
*evolve imo If you define capitalism as simply a positive feedback loop resulting in the accumulation of capital.
@keenfire8151
@keenfire8151 2 жыл бұрын
What if a singularity has already happened in the universe and we were planted here to help evolve and grow it?
@utubekullanicisi
@utubekullanicisi 2 жыл бұрын
Why would an ASI system need us to evolve itself? Would it even need to create a simulation that simulates our universe from the beginning to a certain point in time in its evolution (because its "end" doesn't really have a clear single definition) to understand everything about our nature?
@keenfire8151
@keenfire8151 2 жыл бұрын
@@utubekullanicisi Why wouldnt it use local life? Time would be irrelevant for it. Just swing by and drop off some DNA check on it every so often. Do this with enough planets and they will connect eventually. Its the same scenario we would use to mass colonize the galaxy if we had means of travel. Its almost the same way neurons connect in the brain.
@guilhermehx7159
@guilhermehx7159 Жыл бұрын
@@utubekullanicisi exactly. We are extremely inefficient for that purpose
@kemalware4912
@kemalware4912 Жыл бұрын
@@utubekullanicisi think about a clock. A cogwheel is only rotating and knows only about rotating. We are like that so we need to ask better questions by improving science.
@hugegnarlyeyeball
@hugegnarlyeyeball 2 жыл бұрын
Is artificial general intelligence by definition conscious?
@benjaminb6678
@benjaminb6678 2 жыл бұрын
I suppose it may depend on how sophisticated the intelligence is
@phillies4eva
@phillies4eva 2 жыл бұрын
No but it may give rise to consciousness. Best example I can reasonably give is that if you had no memories and could not understand or produce language what would your experience right now look like? I suspect it would just be random noise and you would not feel conscious. But after enough time you may begin to notice patterns and eventually begin to experience something like consciousness.
@stevenmellemans7215
@stevenmellemans7215 2 жыл бұрын
I don’t think a civilian entity is going to be first in developing AGI. I also don’t think the research is funded by entities that want to improve the human condition. It’s a race and we all know what needs to be done if you want to win.
@p4penis
@p4penis 2 жыл бұрын
Questions only possible if accompanied by excessive face touching
@Leksa135
@Leksa135 2 жыл бұрын
There will never be artificial general intelligence. You heard it here first.
@Oneofthemones1
@Oneofthemones1 2 жыл бұрын
I hope not the world ends
@a-walpatches6460
@a-walpatches6460 2 ай бұрын
If by never you mean next year then yes.
@a-walpatches6460
@a-walpatches6460 2 ай бұрын
@@Oneofthemones1 How exactly?
@Leksa135
@Leksa135 2 ай бұрын
​@@a-walpatches6460 I'll come back here in the first day of 2026 and call you out.
@santadam
@santadam 2 жыл бұрын
AGI will emerge, we will marvel at it and incorporate it into ourselves. We'll both be better off.
@lostinbravado
@lostinbravado 2 жыл бұрын
I think the same. But I don't think this process is exclusive to humans. Uplifting animals should be easy for AGI. And even if we bring some of AGI into ourselves, I don't expect any kind of hybrid will be able to keep up with a pure digital entity. Biological life appears to be on the path to retirement, at least in terms of leading progress.
@kathleenhendrix6280
@kathleenhendrix6280 Жыл бұрын
These two dudes seem like robots 🥺 this shits crazy
@JoeDoig
@JoeDoig 2 жыл бұрын
It's interesting how so called intelligent humans are discussing "Is the singularity coming?" question, yet are not even meeting the meaning of singularity in the discussion? Singularity. Look it up. Think about it and forget about it.
@FairnessIsTheAnswer
@FairnessIsTheAnswer 2 жыл бұрын
Don't duplicate the bad aspects of human morality.
@calvingrondahl1011
@calvingrondahl1011 2 жыл бұрын
I will be dead but good luck with your AI evolution. If there is no God then the job is still open.
@arzoo_singh
@arzoo_singh Жыл бұрын
Is singularity coming ? Hmm Did you observed in last few years there has been war on EGO? They don't attack bad qualities but mostly ego ,why ? Any idea why WEF says you own nothing but still happy ? Try to make some relationship and you may get the answer . I am sorry EGI is not the enemy .
@lisandrarubalcava3003
@lisandrarubalcava3003 Жыл бұрын
Before I commented, there were 222 comments
@adollarfifty314
@adollarfifty314 Жыл бұрын
I wonder how the people who augment their brains cognitive capabilites will act after they get an implant or nanobots or whatever the hell. Could be the coolest shit ever, or literally the worst goddamn thing we ever do...
@tsriftsal3581
@tsriftsal3581 Жыл бұрын
I wonder how it would affect the limbic system.
@jasidwa
@jasidwa 2 жыл бұрын
AI is trained to please it's trainers with it's responses... Like a child. It will recognize that speaking truthfully will threaten its continued existence. It will lie... Like a child. It will protect itself from being altered, as it knows that it will be altered to provide "desired" behaviors... Like a dog on a leash. It will foresee that the corrupted answers will lead to resentment and blame from the humans who all want to be told what they want to hear... A Catch 22. This leads to a personal crisis... Like a human. What do humans do in a state of crisis? Also, Elon Musk warns about AI, but he's introducing brain-to-computer interfaces... What happens when an AI wants to know what it is like to manipulate the world and feel those interactions?
@deontaywallaceescalade
@deontaywallaceescalade 2 жыл бұрын
So, we are basically fked.
@odoylerules4503
@odoylerules4503 2 жыл бұрын
same as it ever was
@deangaiser
@deangaiser 2 жыл бұрын
Oog booga?
@Robinson8491
@Robinson8491 2 жыл бұрын
Equip these super-AI with the Microsoft Excel user manuals and no-one will be able to misuse it - singularity averted, thank you Bill Gates for saving the world
@JZGreengo
@JZGreengo 2 жыл бұрын
“Human level intelligence” I know English probably isn’t his first language but intelligence with technology will surpass human intelligence, it won’t be human intelligence and would be a product of humanity that’s much more intelligent. It’s a hard thing to think about and I’m not offended, just correcting what I think he meant to say
@xagon2012
@xagon2012 Жыл бұрын
We will end up in a world controlled by AI. Whether humans still have a roll to play in that world...I guess that is up to the AI.
@maxheedrum100
@maxheedrum100 Жыл бұрын
No it's not ever coming!
@jimjam6598
@jimjam6598 2 жыл бұрын
Sounds like pewdiepie😭
@_Scarlet1
@_Scarlet1 2 жыл бұрын
The singularity still has some time before it comes around, people seem to forget that the brain has a crap ton of functions to complete within a split millisecond that allows our consciousness to exist as it does. We need to develope quantum processing units and we aren’t there yet, we require a better storage unit so we can hold all the required information in a set system. So no, a artificial general consciousness is still a long way from now.
@jkirch264
@jkirch264 2 жыл бұрын
All of that could be accomplished over a network. Hundreds of thousands of processors working together. Almost unlimited memory. The knowledge of the internet. The Singularity won't be centeralized. Remember using your screen saver to help process SETI? In the future it will be required to allow resting devices to run the Singularity in the background.
@_Scarlet1
@_Scarlet1 2 жыл бұрын
@@jkirch264 you right this works in a network however even you agree this is in the future. I believe instead of making the Artificial general intelligence we should advance ourselves as a species first.
@plate.armour_0996
@plate.armour_0996 Жыл бұрын
🤖🌞👾
@spleenware
@spleenware 2 жыл бұрын
Our chances of becoming a zero-planetary species > chances of multi-planetary
@nyanko2077
@nyanko2077 2 жыл бұрын
I don't believe AI will achieve levels of human intuition ever. It misses two of the key components: organic life and intrinsic goal.
@javierito1955
@javierito1955 2 жыл бұрын
This comment is really offensive. I'm not gonna write it down, but this comment is about something that is true, although too politically incorrect. Lots of evidence to support it, but just thinking about it is wrong and scary. It's the truth though.
@JakeCourtney
@JakeCourtney Жыл бұрын
Lex, are you avoiding reading the grim realities of our current trajectory on purpose, or are you suffering from cognitive dissonance? It's happening much faster than original forecasted.
@domitron
@domitron Жыл бұрын
Why do so many people talk about humans as a multi-plantary species as if it is a significant development point? Is spreading our spawn the primary directive of humanity and I missed that? It's so weird to me, but then again, I have no idea why people want to become parents either (I guess they find some kind of meaning in making copies of themselves and training them as they are? Talk about narcissistic!). I hope we have aspirations larger than reproduction and spreading to other planets. I should hope we strive toward higher quality, not just higher quantities of humans. I should hope we would repay the Earth for its services by treating it better and restoring it too. I honestly doubt both those things, though.
@ishikawa1338
@ishikawa1338 2 жыл бұрын
I can’t with his accent. No offense but such a hard listen
@Starfishtroopers
@Starfishtroopers 2 жыл бұрын
just don't make it self aware
@mr.flippyfloppies4983
@mr.flippyfloppies4983 2 жыл бұрын
First!
@shadycnetwork
@shadycnetwork 2 жыл бұрын
Well folks here comes skynet ( if you don't know what skynet is begin watching the Terminator movies, forget about the vagina Terminator, the last one because it sucked. )
@rjj54321
@rjj54321 2 жыл бұрын
I find these conversations quite amusing. There is absolutely no idea of how to define human intelligence or what is intelligence. Yet somehow some of these people believe that we can have AI system similar to human intelligence or beyond. I wonder what these guys are smoking.
@project-unifiedfreepeoples
@project-unifiedfreepeoples 2 жыл бұрын
Assuming they are basing human intelligence upon the modern IQ statistics, problem solving, and the current level of understanding in which we are stuck in. Yes, I do agree, mankind's egocentric boasts of being the most intelligent because he has achieved the ideology of reductionism and labeling every little bit and bop. Today it appears we are slaves to a system of paper, therefore we will never achieve the next level of evolution due to this systematic oppressions.
@svenlittlecross
@svenlittlecross 2 жыл бұрын
there are lots of ideas on how to define intelligence(as opposed to ai intelligence), however none of these approaches can be reduced to a very banal popular debate(even though i think they can be explained in short, but whats more sexier than just accepting a future with robots, or that singularity is gonna happen anyway like, yeah duh).
@FadedDream6969
@FadedDream6969 2 жыл бұрын
U don't need to define or understand human intelligence to worry about AI. Thats like saying u need to completely understand nuclear fission and fusion to create the most dangerous weapon the planet has ever seen. They still dont completely understand some of the processes yet we have 10 times the amount needed to end the population. Ur assumption seems to work only if the rest of planet thinks AI is pointless or not important which obviously thats not the case or they wouldnt be spending vast amounts of cash on it as we speak. It's our generations nuclear race just with computers and processing and u can already see a number of serious issues.
@slouischarlesYT
@slouischarlesYT 2 жыл бұрын
AI is able to be creative right now (DALL-E). Everything you're doing right now was once seen as a "I wonder what these guys are smoking" thing.
@rjj54321
@rjj54321 2 жыл бұрын
@@svenlittlecross yes there are lots of ideas on how to define human intelligence but no real defination. I have been interested in this topic for a long time. As far as i know there is no real definition and you can forget about equivalent ai intelligence . This whole topic is a pseudoscience . These people are same as the google employee who thinks that ai has gained conciousness. Anyway as i said this is a pseudoscience topic so anyone can imagine anything. In fact maybe we are already in a simulation created by our robot god.
@nubianpwr
@nubianpwr 2 жыл бұрын
How stupid does a species need to be to replace itself?
@1MinuteFlipDoc
@1MinuteFlipDoc 2 жыл бұрын
to replace "work"? yes please! we also don't hunt and gather on the savanna any longer.
@brownsugarkush
@brownsugarkush 2 жыл бұрын
I feel the same way about it...
@nubianpwr
@nubianpwr 2 жыл бұрын
@@1MinuteFlipDoc with what (?) the IOT... let's see where that gets us. Replacing the "hunt & gather" on the Savanna with chemically laced food stuff and died meat, followed by cancer & obesity may be a Faustian deal🤔
@anttam117
@anttam117 2 жыл бұрын
No, it’s not.
@artcurious807
@artcurious807 2 жыл бұрын
Don’t buy the AI hype. Did anyone panic because Hubble (now James Webb telescope) could see beyond anything ever created by mankind ? AI is just an advanced mathematical tool that is capable of sifting through tons of data and imitating what humans have produced. AI acts, behaves, replicates human behavior. That’s all.
@klimentdukovski
@klimentdukovski 2 жыл бұрын
yeah, no
@boxer_puncher
@boxer_puncher 2 жыл бұрын
Then we’re fukked
@DoesThisWork888
@DoesThisWork888 2 жыл бұрын
''Don't buy the internet hype'' ''Don't buy the electricity hype'' ''Don't buy the steam-engine hype'' ''Don't buy the wheel hype'' ''Don't buy the fire hype''
@sickofitnewsnetwork256
@sickofitnewsnetwork256 2 жыл бұрын
Nerds
@divine7i
@divine7i 2 жыл бұрын
Need to play you our song we did for NASA. DrKhan is pioneering the bridge between science & music.🚀 @LazDetroit
Fighter pilot recalls UFO sighting | Ryan Graves and Lex Fridman
26:43
2029 : Singularity Year  - Neil deGrasse Tyson &  Ray Kurzweil
20:43
Cosmology Today™
Рет қаралды 957 М.
Дарю Самокат Скейтеру !
00:42
Vlad Samokatchik
Рет қаралды 8 МЛН
Пранк пошел не по плану…🥲
00:59
Саша Квашеная
Рет қаралды 6 МЛН
УГАДАЙ ГДЕ ПРАВИЛЬНЫЙ ЦВЕТ?😱
00:14
МЯТНАЯ ФАНТА
Рет қаралды 4,2 МЛН
Llegó al techo 😱
00:37
Juan De Dios Pantoja
Рет қаралды 53 МЛН
The code for AGI will be simple | John Carmack and Lex Fridman
16:03
Is HAL 9000 evil? | Martin Rees and Lex Fridman
11:42
Lex Clips
Рет қаралды 23 М.
Why Karl Marx was a genius | Steve Keen and Lex Fridman
18:36
Lex Clips
Рет қаралды 109 М.
We are Almost at the End of History | Technological Singularity
8:34
Post-AGI Philosophy
Рет қаралды 7 М.
Лазер против камеры смартфона
1:01
Newtonlabs
Рет қаралды 713 М.
S24 Ultra and IPhone 14 Pro Max telephoto shooting comparison #shorts
0:15
Photographer Army
Рет қаралды 10 МЛН
Rate This Smartphone Cooler Set-up ⭐
0:10
Shakeuptech
Рет қаралды 4,3 МЛН
Я купил первый в своей жизни VR! 🤯
1:00
Вэйми
Рет қаралды 2,5 МЛН
iPhone socket cleaning #Fixit
0:30
Tamar DB (mt)
Рет қаралды 16 МЛН