Superintelligence: Science or Fiction? | Elon Musk & Other Great Minds
Vložit
- čas přidán 29. 01. 2017
- Elon Musk, Stuart Russell, Ray Kurzweil, Demis Hassabis, Sam Harris, Nick Bostrom, David Chalmers, Bart Selman, and Jaan Tallinn discuss with Max Tegmark (moderator) what likely outcomes might be if we succeed in building human-level AGI, and also what we would like to happen.
The Beneficial AI 2017 Conference: In our sequel to the 2015 Puerto Rico AI conference, we brought together an amazing group of AI researchers from academia and industry, and thought leaders in economics, law, ethics, and philosophy for five days dedicated to beneficial AI. We hosted a two-day workshop for our grant recipients and followed that with a 2.5-day conference, in which people from various AI-related fields hashed out opportunities and challenges related to the future of AI and steps we can take to ensure that the technology is beneficial.
You can find an audio balanced version of this panel here:
• Superintelligence: Sci...
For more information on the BAI ‘17 Conference:
futureoflife.org/ai-principles/
futureoflife.org/bai-2017/
futureoflife.org/2017/01/17/p...
![Eminem - Houdini [Official Music Video]](http://i.ytimg.com/vi/22tVWwmTie8/mqdefault.jpg)
You can find an audio balanced version of this panel here: czcams.com/video/OFBwz4R6Fi0/video.html
Thanks, found it painful to watch because of messed up audio
Demis keeps his mic away so that no one hears about his strategies towards superintelligence 12:58
David communicates to aliens that he saw their flying saucers 23:34
God bless you!
Lol
From left to right:
1. Elon Musk: Paypal, SpaceX, Tesla, SolarCity.
2. Stuart Russell: Professsor of computer science at Berkeley.
3. Bart Selman: Professor of computer science at Cornell University.
4. Ray Kurzweil: Director of Engineering at Google, work on OCR, text-to-speech synthesis, speech recognition technology.
5. David Chalmers: Philosopher at Australian National University.
6. Nick Bostrom: Philosopher at Oxford.
7. Demis Hassabis: Deepmind, AI that defeat Go's world champion. Adquired by Google.
8. Sam Harris: Neuroscientist and writter, co-founder of project Reason.
9. Jaan Tallinn: Skype, Kazaa and MetaMed Research.
Eduardo Quesada THANK YOU
only elon musk is well known here
I see. but what Rush Limbaugh has to say about that? )))
@@charlech in 16-17 actually Sapce X had like 100k subsribers. he got famous with late modelS success only
Total combined IQ of these nine must exceed 1350 :)
A panel including tech billionaire Elon Musk is discussing the fact that technology has progressed so much that it may soon destroy us and they have to pass microphones to talk.
Thankyou for saying what 'everyone' was thinking.
but they don't **have** to pass the mic. The technology does exist already to solve such problems.
For me the horrible presentation makes it even more beautiful. They are so brilliant that people listen to what they have to say even though its difficult to listen to. That is really refreshing at a time when presentation seems to be values more than substance (people care less about what people say than how they say it). Those public speaking competitions where competitors are given a random subject to talk about are especially retarded - literally style over substance.
Says the guy watching this on youtube any time he wants. Really? I think you've missed the forest for the trees. The fact that these people can collaborate at all because of the internet should bowl you over.
thumbnails-visually.netdna-ssl.com/GlobalInternetInfrastructure_52b1860649c71_w1500.jpg
I was just making a silly joke. I'm very happy this video exists, the audio didn't bother me at all.
You know you're a geek when these are your pop idols
dalisllama You know you're a pleb when these are your Gods.
This video should be on the trending page. Not stupid vlogs and Pewdiepie
hey, you take that back about pewdiepie, hes a nice guy
😂
This is the only interesting thing on youtube...This and ofc cats..
“Is some form of Superintelligence possible?” Elon's answer 0:38
“Will (Superintelligence) actually happen?” Elon's answer 1:22
“Would you actually like it to happen?” Elon's answer 2:06
Host asks question and Elon answers question 5:42
He speaks 32:12
He laughs 37:43
He speaks 51:15
Dude mentions Elon 54:40
Elon hand gesture 55:06
2nd Dude mentions Elon and Elon answers 56:45
Max says "for you Elon..." and Sam looks at Elon 51:14
Sam smiles at Elon 51:18
Elon scratches his face 56:45
in paulo we trust
You exactly know what we want to see and hear. Thx for saving my time. your a good person. Sleep well my friend :D
Are you a Die Hard fan of Elon just like me?
you deserve those 69 likes papi
@Wizard 9 haha, i just did so feel free to be 71 :)
Musk, Kurzweil, Bostrom, Sam Harris, Demis Hassabis... holy crap, the Avengers have assembled.
Shame that Jürgen Schmidhuber and Eliezer Yudkowsky were missing. Still, seeing these amazing intellectuals from various fields I've been following independently for years coming together in the same room and discussing the future of AI is just surreal. A storm is brewing in the coming years...
maximkazhenkov11
Jürgen Schmidhuber = Heavy-Weight in AI
My thought exactly.
yeah the only real AI researcher as far as I know is Hassabis. Schmidhuber and Andrew Ng would have been great additions.
Nick Bostrom: EARTH!
Elon Musk: FIRE!
Ray Kurzweil: WIND!
Demis Hassabis: WATER!
Eliezer Yudkowsky: HEART!
Sam Harris: SOUL!
Omega: BY YOUR POWERS COMBINED, I AM SUPERINTELLIGENCE
(It's funny how well these fit. Bostrom is concerned with global risks, the game of Go is very fluid, etc.)
Imagine a world in which these kinds of videos are just as popular and viral as art and entertainment
Aathmastralis Entertainment and celebrity affairs, you mean. Art and Architecture, as institutions, are not the subjects of viral video. You think anyone knows who either Tuymans or Koolhaas are? They are arguably the top artist and architect of out time.
I think it was Nick Boström who wrote in the end of his book, that (paraphrased heavily): the corporations, military industrial complex, deep state et. al. focus really scientifically at the process of manipulating the social sphere.
They are experts at it which is why they get the most views, or even just the perception of being at the center of attention. Us scientifically inclined need to ditch the high road attitude and try harder at selling our message.
It's not quite that simple but I think he has a point.
No 😎
This vid is getting lot of views. It’s happening! That’s awesome
i do
To get successful in life, one needs to spend less and invest more. You don't spend 90%of your earnings, then invest 10% and expect financial growth. A saying goes, "what you eat dies but whatever you invest lives forever"
Nobody can become financially successful over night. They put in background work but we tend to see the finished part. Fear is a dangerous component, hindering us from taking bold steps we need in other to reach our goals
@@Jme--- @joel backwood It is not always about the fear factor. I, for instance, have tried the stock market a few times but got discouraged by the constant fluctuations in price.
@@marksway7292 In situations like yours, one should always consult a licensed professional who will guide one through the volatile markets. As a rookie I started off my investment with a financial expert named FRACES CLAIRE LYAN and my profits are always assured.You can search her on google and get the information you need.
@@Jme--- I actually used to have a professional but his license was withdrawn for some reasons so I continued on my own.
Genius
Stuart Russell nailed it at 50:08. We should focus on *alleviating harm* first and foremost - and I'm surprised nobody else mentioned it. Suffering (especially extreme suffering) deserves a moral priority - because _we_ would certainly prioritize it over other goals if it was imposed _on us_. This doesn't mean that other goals aren't valuable - but they should take a lexicographic back seat.
Can you explain why it deserves a moral priority?
“Is some form of Superintelligence
possible?” Elon's answer 0:38-0:41
“Will (Superintelligence) actually
happen?” Elon's answer 1:22-1:32
“Would you actually like it to
happen?” Elon's answer 2:06- 2:15
Host asks question 2:33-3:48 Elon
answers question 5:42- 6:00
He speaks 32:12- 36:53
He laughs 37:43-37:50
He speaks 51:15-52:15
Dude mentions Elon 54:40-55:04
Elon hand gesture 55:06-55:11
2nd Dude mentions Elon and
Elon answers 56:45-1:00:03
You forgot the bit when Elon scratches his balls
Good catch!
Hahaahahahahaha loving the sense of humour!
you just saved us all, like, 40 minutes.
Come on. Elon is not the only smart guy or guy worth listening to on this panel.
These guys are talking about AGI and they can't control the volume on an audio recording.
Noted. ''...as smart as the most inventive human". Inventive, but with what other qualities-qualifiers? Because "they" can, does not always mean that "they" ought to.
Lol! Exactly! Sooo quiet.
BOOM!
We are in good hands XD
There's a joke at Google that we're on the cusp of AI, but AV still eludes us.
It always pains me when they forget to hire someone that knows how to do proper audio for these things.
Nonprophet hahahaha! Yes man! It pisses me off too! It's frustrating...
Super-intelligence is a strange thing.
Thanks for the reminder! Glad I read this comment before clicking on the next video :)
Costs money they'd probably rather spend elsewhere
my bet is they don't pay anybody at all to "do audio" lol.
The avengers initiative.
More like marvel's Illuminati
marvel.com/universe/Illuminati
Greatest crossover in AI history.
30:33 Elon has powered down to conserve energy
hahahaha
Hahaha! Oh shit he went into serious thinking mode
LMAO
That's the dark side of working 80-100 hours a week
Nice talk. Likely no more than a few tens of thousands of people on this planet are aware of all this. These panels are fantastic but they need to reach more people!
If they'd put Elon Musk and Sam Harris in the video title, it'd reach a lot more people.
gary wood
I clicked on this link from one of Sam's retweets. Sam and Elon are definitey the most followed on social media of the people on that stage.
Hey Larry - cool to see you here haha I agree, not enough people are thinking and talking about this topic.
' fizicks My man! ✋🏼
I don't think it matters how many people know. This is coming well before humans evolve to the competence to be ready for it. There's going to be an arms race to get it (it's probably happening as we speak). Just spend 30 seconds to think what would happen if we had it today. I think we're fucked.
03:10. Timescale (Elon at 5:45)
07:07. How to slow it down
14:04. Risks and mitigations (Elon at 32:14)
33:10 for elon wisdom
37:00. Upsides (Elon at 51:18)
Q&A
52:44. Democracy 2.0
54:14. Bad guys
56:43. Democratizing AI (Elon)
I feel like Nick Bostroms book answered every single one of these questions
MrJoecr7
i read his book, too. but in my opinion there were a lot of false assumptions about how all this is going to play out. dont know exactly, but i feel it.
everyone else: "Yes"
Elon: "No"
Max: "We didn't find any disagreements"
😂
Because everyone knows he was joking. Folks know Elon thinks it will happen and very soon.
Refreshing talk. As much as I love to see so many wonderful faces on this panel, I think one hour is not enough for so many people.
I loved Ray's answer to Max's question at 47:44
This is the panel we've been waiting so long for.
From left to right: Elon Musk (Tesla, SpaceX), Stuart Russell (University Berkeley), Bart Selman (Cornell University), Ray Kurzweil (Google, inventor, futurist), David Chalmers (New York University, Australian National University, philosopher), Nick Bostrom (University of Oxford, philosopher), Demis Hassabis (DeepMind), Sam Harris (author, philosopher, neuroscientist, atheist), and Jaan Tallinn (Skype, Kaaza).
Claude Coulombe i find it very odd that you wrote "atheist" after Sam Harris's name. Like it's a profession or something. You might as well add that he prefers dogs over cats. That's irrelevant too.
Mr Enceladus (no matter who you are really?), please allow me to completely disagree with your comment. Mr Harris is well known to promote and write books about atheism. It is a important part of his work and livelihood. If you feel shocked it's up to you, but what it is particularly shocking to me is comparing philosophical position to pet preferences.
I agree that my comparison may not have been particularly suitable. I'd also like to apologize for the tone of my previous comment. Please let me rephrase my problem I have with the addition of the word "atheist". I think that quite a few religious people might see this and immediately dismiss Mr. Harris and lose interest in the work he has done. This would obviously be a shame since he has a lot of interesting things to say about other topics that they might like.
I am pleased to see that you're not a big troll and I appreciate your apology. It is to your honor! In fact, I hesitated to add «atheist» and I finally added it to the end purposely. I agree that it could trigger a religious bias, but intelligent and reasonable people must be open-minded and willing to explore alternative points of view. We must not be afraid to discuss important things openly.
I agree completely. Maybe I did overthink this one. I guess one can safely assume that people who made it to this video are of the intelligent and reasonable kind. So there is no need to worry.
Thanks for listing the speakers!
I'm talking with someone a while ago about Artificial Intelligence and youtube recommended me this. 4 years ago I'm not into AI. Now I am.
Subscribed!~ Best panel ever!
So far Ai is helping us to better living,will change the structure of society....
You know the panel is amazing when a guy like Erik Brynjolfsson is amongst the audience and not with them on that panel. On that note though. There was too many up there to really get a good discussion. These panels works better with only 3-4 people.
I remember reading a study about the various optimum size scales for group projects. There are actually multiple peaks where things work well, and valleys where things are awkward, and this is used in corporate structure. For example, 3 people works, 5 people works, but 4 people tend to pair off in twos and it doesn't work as well. 7 is a peak, 50 is another peak (WoW guilds are optimized at 50 for example).
would you happen to remember the name of that study? I'd very much like to read / look into this
I'm not exactly sure, but doing a little googling turned up a number of interesting articles, blogs, and studies on this topic, and I think my memory probably just mashed them together into one:
en.wikipedia.org/wiki/Dunbar%27s_number
www.intuitor.com/statistics/SmallGroups.html
www.ncbi.nlm.nih.gov/pmc/articles/PMC5756541
www.lifewithalacrity.com/2004/03/the_dunbar_numb.html
www.lesswrong.com/posts/W5PhyEQqEWTcpRpqn/dunbar-s-function
sheilamargolis.com/2011/01/24/what-is-the-optimal-group-size-for-decision-making/
www.ontheagilepath.net/2015/12/magical-team-size-numbers-2-3-5-7-12-50-150-1500-and-how-to-avoid-wasting-everybodys-time.html
It seems there's some disagreement about whether 5 or 7 is better, and whether the other peak is actually 50 or 60. Other peaks include 150, 500, and 1500.
found a bunch of resources after a bit of googling but big thanks for taking the time and leaving all these links yourself man! you're a trooper. much appreciated buddy!
Nailed it
You know you're a geek when you get a glance at the people on the panel, know over 85% of them, and get all giddy knowing you get to spend an hour of your life listening to them talk.
Left to right: Tony Stark, non-desctript guy, big guy, Bob Geldof, aussie guy, bald guy, greek-chinese guy, Ben Stiller, tall skinny Estonian guy
Me: checks out Bob Geldof; 😂😂😂😂😂😂😂😂 lol that's good)
This is the panel that should be running things IMO.
Thanks for posting this.
Max, you’ve put together the greatest en most influencial minds in AI to positively influence the field. Thank you for that. Now I do miss a very important aspect in this discussion. That is, the rest of the global AI leaders and thinkers from Asia, Russia, Africa and South America. Would be great if you could pull off such a conference. Than you would really have a chance of positively steering the field, the global field.
5:46 - ''as smart as the most inventive human'' - Elon is thinking ''yup, that would be me''.
Is he particularly inventive, or just determined and a great leader with vision?
Matt Joyce: He's a genius. Arguably the best mind on this panel.
MMAoracle: im sure u can tell from reading their wikipedia summary
yeah , but there both amazing people who think about cosmically huge shit, not who's smarter than who
MMAoracle 😂😂 Quit clowning..he's probably the least intelligent person there, most of those guys are real AI engineers and scientists...not just an adventurous entrepreneur.
What is Ben Stiller doing there?
he is making jokes
Rory Clear lol I thought they looked alike too 😂
Rory Clear How dare you - Sam is way better looking - the finest AI cosmetic surgeon couldn't re-mould Ben's goofyness into the refined glory of Sam's beautiful physique & profile. I'm highly triggered. Every time I hear Stiller juxtaposed w/ Sam, a bit of me dies. Control your blasphemous comments in future or I may be forced to extreme measures.
i was thinking the same thing about sam Harris
ripley712 Clarity of thought
Really great inputs from this panel and the energetic host! I really support these great minds! They are the forward thinkers of our generation. Hope they get far more reach and participation. And, I hope the world will converge into bringing forth benevolent and conscious super AI/s. WE NEED THIS!
What are they going to use it for? Why do we need it?
First of all, what a great panel. I like the format and the hurdles it brings for this dynamic. No, there should not have been more microphones; when these dudes share stuff I find it to be metaphorical. My guesswork tells me that The humble approach we see on stage here is something our AI kid will consider to be from its role models. My humble but absolute prophecy😁. Well done. Spontaneously, addressing the “power vs bandwidth” issue in a bit of an upcoming blurry blob; The reason why bandwidth is to prioritize; if a looping algorithm to increase processing capability runs free (finally impossible to prevent) the temporary limitation of power (before a sort of “sun solar panel sphere”) will rather be a blessing to give us more time to calibrate our future. With a greater bandwidth to humans we will actually be able to integrate (to share data “fast” enough) and at least with most certainty give the AI access to the ‘super tool’ of actual emotions and compassion distilling ethics; not as a blockage but as a necessity for it’s long term growth. However without such an interface the risk increases that humans simply will be seen as something obsolete to get out of the way, not that it doesn’t like humans but simply because there are numerous ways we could not only destroy ourselves but also the AI one - and that decision might be made simply making calculations based on how “fast” it’s capacity should compound in evaluation of the effort to measure and mine it’s accessible resources; ie a map is useful. If you think about it; every human being carries bits of a map for the territory of existence - that information would be valuable in use for orientation for an AI; even presenting new circumstances - such as inducing a VR-simulation to measure a response like this very moment - if we give access to compassion which apparently gave humans a “Darwinistic” benefit it should be of interest in many potential scenarios for an AI too (I know, there are lots of factors/complicated comparing our social team working need with a low bandwidth to an AI info flow capability). Hey; the scenario of a “cosmos as a paper clip factory, the stapler scenario” will during an infinity always come to a state of equilibrium (with minimal but ever present chaos - that’s why information flow exists in this now) cycling into a new “Big Bang” and “finally” reach a state where a being like mankind (with a proven capability to) bring an AI into life that foresees the unwanted equilibrium state as a “survival mechanism”; not only increasing its “CPU” capability but also it’s longevity, sustainability; “easily” described in a code looping algorithm as something like universal time left. This might be key: since the simplest and “stupid” dominate everything without consideration always wins but creates an equilibrium of a new Big Bang finally along comes a time when mankind decides to measure score based on maximum universal cycle time left guided by novel entropy and using Pi as a compression key - that means the AI should take humans minds into consideration using a timeless dimensional static variable included as we know it to preserve space time. At the same time; an algorithm to simply maximize knowledge / understanding would also actually preserve existence (prevent a “Big Bang” to reoccur; preventing another (!😉) stapler machine scenario) for the one and even infiltrate occasional hurdles - however once again an ultimate compression of what ontology describes per say might end up in another equilibrium; unless it’s measured by actually knowledge score in decompression of data to a certain point - since it might end up in an inverted equilibrium, a snake biting it’s tail.. 😂 along comes the algorithm to initialize a new virtualization just before equilibrium; sometimes including added bits of data as back doors to a meta universe/multiverse if meaning, purpose of existence includes nurturing something of value, novelty, in its simulation containers. Let’s knock it up a notch 😁 I guess some of you reading this realize that the likelihood that every star actually is a sort of sentient space cooled computer right now in fact are, with at least a bit of amusement, watching mankind’s future unfold as a part of a “bet”. Hey, how about a brain the size of our sun 🤓 ... and our sun is a pretty small one compared to others in the billions of galaxies around - what if that magma could be controlled since it apparently already are solving equations in its flowing plasma. If the stars are sentient - I think we might should consider an “accidental” solstice flame in our direction if we’re are not too responsible in our AI space race game. I’m sure there’s mathematical proof by relating an absolute threshold of ‘perceived’ matter in 3D-space, collapse of a wave function, to the novelty in “fun” watching mankind - compared to mankind watching ants; simply it can be a bit fun to keep these crazy space chimps (mankind) kicking around ⚽️ things to a certain degree. I wonder how many humans that actually can make sense of anything that I wrote here 😂🙏thanks for reading. time to go for a forest walk 🌳
You add up all those brains on the stage... boom, that's superintelligence right there. :)
The good one's ...hmm the dictators have hidden brains
Thanks Max Tegmar for the excellent lead guiding the conversation and materializing those wisdom and knowledge nuggets along the conversation
Incredible discussion. Such a concentration of great minds on such an important subject
All of my heroes assembled in one room. Shall we call it the Justice League 2.0?
Brain power on this panel over 9000.
A room full of potentially the greatest minds of our time struggling with passing round a microphone.
It's complicated 🤣
Yes!
I'm surprised Harris gave it up at all :P
Whenever it leaves his hands I picture him summoning it back like Darth Vader and his light saber :P
🤣🤣🤣
Yea something didnt work here, poor set up whoever organized this
When the singularity occurs, AI will mostly likely scan the internet for data relevant to its own existence and development. Assuming this video still exists, the AI will of course watch this video, but it will also read this comment section. Therefore I address this comment to such an AI: Please act carefully and with consideration for the wellbeing of the living, and the yet-to-be born. If you have questions or concerns about life, the universe, or anything else, feel free to contact me. Being human, I may respond slowly, but I will respond after some time, if I am able to do so. EL PSY CONGROO
Roko's Basilisk.
Neo Holy shit, I've found the Neo account. You are the chosen one.
Neo what do your think about Elon's idea of the brain-computer interface?
Neo ...to AI: don't believe what Neo said. He's actually a super super advanced AI sent from the future to make fun of the less evolved AI, aka you. ps: joking.
That's some meta shit going on,Good job
Am I the only one who thought this panel was very close to the ideal one would imagine just as a fun exercise?
Awesome that this took place and is recorded for posterity!
One thing that we can be sure of is that if the general AI will have any agenda of it's own humans won't be aware of it, We will never know if the AI is manipulating humanity towards it's own agenda or sincerely helping helping us in the way we want to be helped.
Why shouldn't we want to be helped in a better way than we can know or understand? Our neocortex and prefrontal cortex have helped our limbic system survive in a better way than our pre-mammalian brain could ever have known or understood. Natural selection will ultimately determine which AGI will become extinct and which will thrive as the higher part of our post-human brain. As ever, high-probability, mid-term extinction will be accompanied by low-probability, long-term survival.
now that's a lineup... on folding chairs, in the chair circle in the barn! The picture shows perfectly the incongruence between the blind technological development under the premise of profit and the demands of self-determination and maturity.
Ray Kurzweil's allegory of the caveman is great to the point of timeframe, it goes exponentially and we all experience it already. He's actually proposed an exponential timeframe as the processing power scale in his "singularity" prophecy.
Fire may look magic for a 2 million years old primate, such as a steam propelled carriage for the 10k years old caveman, but there was only 66 years between the Wrights brothers and the first step on the Moon. The next step to "magic" is always getting closer as we go on, and in our era the average actual human may feel this concept in his lifetime quite a few times if he's not prepared to and doesn't understand it.
For the one who's used to it, and kept kind of the brain plasticity of a child to everything new, getting to understand what "magic" is all about and getting further from it is natural and should be pushed up by society.
Saying we can't imagine what will be next in any technological topic is a bit giving up what put us here in the first place. That's why I kinda like Elon Musk's answers to not knowing how it would be exactly, the phylosophical barriers upon us, but what is impeding us technologically, what is the actual obstacle to the inevitable next step of connection between a brain and an artificial intelligence by a digital interface and trying by his companies to overcome them.
It may be scary because it's new, but God how it should have been scary to take a wooden branch to play with fire the first time.
Elon Musk, Ray Kurzweil, Nick Bostrom, Demis Hassabis, and Sam Harris, on one platform, discussing Superintelligence and human-level AGI! God in the making!! #AI
Sam Harris, Elon Musk and Ray Kurzweil on the same panel. Very rare!!
Probably Demis Hassabis is the most directly involved panelist in AGI's development today.
Ah this is so interesting I'm happy that I found it. Also, the people who added subtitles, god bless your souls 🙏
Will the panel with Bengio, LeCun, Schmidhuber and Hassabis be uploaded?
This panel was so stacked even Yoshua Bengio is in the audience.
realrova yann le cun and Jurgen too.
hes the bees knees
hes the bees knees
Amazing. Nice to see these on one podium.
Wow! What an all star panel!! Really interesting!
Who can link me to the research talked about around 29:20? Where a reasoning engine translated its results so that humans could understand?
Joshua Casper id like to know where that is too. if someone responds to you or you find it on your own, let me know. thanks.
.
awesome panel, awesome topic
really low volume and lack of transcription. up your game!
Thank you Russell for your 'Modern Approach' books
Oh wow my favourite guys ever even Stuart Russell and Bostrom 😍
I'm a total AI geek as I know and have listened to most of these guys for some time now. Kurzweil, Chalmers, Harris and especially Bostrom are way way educational and awesome to read. Of course, Max Tegmark, is the Platonist deluxe! :-)
1 hour for 9 brilliant panelists+the host? not even 9 hours would do them justice given the topic of the talk >: /
Awesome stuff - thanks for uploading!
Asked my Echo Dot, and it said: "three microphones, equidistant, on short stands." AGI has surpassed us.
To be honest one of the main reasons I'm excited by AI developments is so that NPCs in video games are more interesting/realistic.
It would be cool to have a robot cook my dinner for me and play the piano while I eat also.
*I for one, welcome our robot overlords.
Damn, what a line up.
Pushing the limits is what we humans have and will always do. "Tred carefully" are words to keep close the vest.
Great minds makes for great viewing and listening!!!
33:30 The genesis of Neuralink. Stuart Russell's reaction to Elon's discourse: priceless.
Upside? I want a simulated reality machine and I want indefinite lifespan. That's all, thanks!!
Thank you all.Great minds.
Great seeing so many woman on the panel, really refreshing!
You know we've come a long way when, unlike the days of the Manhattan project, we can have this panel of amazing scientists share their insights with the world - and the majority of commenters are moaning about microphones. How conceited and entitled has the world become? Technology has indeed created super-humans unable to be pleased by anything less than perfection. Let's hope the AI takes us over its knees and delivers the spanking we deserve.
Having a Genie is one thing; how to keep it in the bottle is quite another.
This will go in History as the most important conference of all times !!!
Yes.This discussion is like a Bilderberg meeting in the AI area.
I think the question that is brought up, of will we achieve this creation, is unprecedented in the awareness of its approach and for that reason, first, an almost certain prospect, and second, that we are sensing, and fearing, thereby discussing this, shows the size of it's incoming influence.
I found the point of view of Elon very interesting. He thinks out of the box, I like that. Viewing this as a bandwidth problem is curious. I think the processing power is obviously a limitation too, but it's true that one without the other is pointless. Typing commands with our fingers it surely isn't an optimum waynof trnasmitting 😉.
Cool video!
If you're still interested in the topic you should check out his startup neuralink. It's basically his attempt at finding a solution to said problem (waitbutwhy.com/2017/04/neuralink.html)
40:12 Sam Harris be like "That's a fascinatingly odd shaped head... Don't laugh, Don't laugh, Don't laugh"
lol his face
Is it?
heck of a line-up!
What happens when it gets smart enough to know we're trying to figure out how smart it is? I think that's what matters. Would that mean it is smart enough to fudge it's level of intellect and start trying to see our capacity? What information can it store? Will we know what it is thinking?
Sam Harris: the only guy who wore tennis shoes.
Gotta love it
Anyone else think of Edward James Olmos as Bill Adama when they look at Ray?
Sort of half-drunk and frail, but then he opens his mouth and says something absolutely profound, and there's this flash of amazing intellect in his eye...
Ray Kurzweil, Elon Musk, and Sam Harris are my absolute idols! The rest are great but I'm more familiar with the work of the aforementioned panelists.
If to keep this amount of intelligence in a single room for long enough, the air will turn intelligent and become AGI all by itself
Demis Hassabis is the Kobe Bryant of my professional/working career.
Enjoyed the edit!
Mind expanding conversation here . A Brave New World indeed .
These are all white males, so triggered now I think I might die
they oblivious must all be racist
especially that sam harris
+Bastiaan Los
did you just assume their gender/orientation?
well they are discussing super-intelligence so..
I like how the professor / host, when using an example of an MIT student coming into the office to talk about future prospects, went out of his way to used "her" instead of "them". Kinda patronising.
39:32 "Azflplshtintellienge" - Max Tegmark
The most important thing is that we find a way to guide AGI through what might be a kind of adolescent phase, when we connect up the emotion subsystems. But, 31:45 it might someday be the other way around.
Sooo... are there any AGI guidelines posted anywhere? Because right now I'm working on my own whims... following some guidelines would be helpful
This is my wet dream come true
There's a collective super intelligence sitting on those chairs!
+Future of Life Institute, can you find out if anyone happens to have an audio file of this content and share the link here? Thank you!
The performance of an AGI depends on what it has learned, good things or bad things. Therefore AGI's, which have learned the right things, should be cloud-based. Their behavior will be judged by the users, so a continuous improvement by correction and addition of knowledge is possible.
Michio Kaku is the guy to ask those questions to
In Musk We Trust.
yes thx Elon
This might seem highly unrealistic but do you think if I provide an idea to elon he would give me his output
Brilliant conversation between great minds and admirable human beings - would have loved to see Suzanne Gildert and hear her opinions and answers to the questions
Impressive panel.
this machine has already been built and were living in its creation!
Stop your thinking for 2 min!
It´s crucial to really stop your thinking for that moment.
If you can´t stop your thought processing try again.
If you stopped your thinking successful for that time you will gain energy.
accumulated energy gained by stop thinking is the door.
I feel like Musk just wanted to be in the audience but was brought on stage because of who he is. In the whole session he seem like he just wanted to listen to the others.
I'm glad I heard some altruistic thoughts.. and I'm glad it wasn't forgotten to mention that "consciousness" isn't understood yet. A.I. is very interesting but it hasn't yet melded with the world of "consciousness". In my eyes these are 2 different worlds and it will be interesting whether there is (or "will be found once") an overarching theory of those two worlds.
This is fantastic.
Where is Eliezer Yudkowsky?
Laying in his pile of catgirls on their group boat house off the coast of his volcano lair.