Noam Chomsky on AI: The Singularity is Science Fiction!
Вставка
- Опубліковано 18 чер 2024
- www.singularityweblog.com/noam...
Dr. Noam Chomsky is a famed linguist, political activist, prolific author and recognized public speaker, who has spent the last 60 years living a double life -- one as a political activist and another as a linguist. His activism allegedly made him the US government's public enemy number one. As a linguist he is often credited for dethroning behaviorism and becoming the "father of modern linguistics" (and/or cognitive science). Put together his accomplishments are the reasons why he is often listed as one of the most important intellectuals of the 20th century. And so I was very much looking forward to interviewing him on Singularity 1 on 1.
Unfortunately our time together was delayed, then rushed and a bit shorter than anticipated. So I was pretty nervous throughout and messed up some of my questions and timing. Never-the-less, I believe that we still had a worthy conversation with Dr. Chomsky and I appreciate the generous though limited time that he was able to grant me.
During our 28 minute conversation with Noam Chomsky we cover a variety of interesting topics such as: the balance between his academic and his political life; artificial intelligence and reverse engineering the human brain; why in his view both Deep Blue and Watson are little more than PR; the slow but substantial progress of our civilization; the technological singularity... - Наука та технологія
![Singularity Or Bust [Full Documentary]](http://i.ytimg.com/vi/owppju3jwPE/mqdefault.jpg)
This Nikola guy does a very nice job with the interview- he asks questions and then gives time for detailed responses without interruption. Well done.
+James Nation +Nikola Danaylov Agreed!
+James Nation absolutely, makes you wonder why people who are paid to do that by CNN and Fox news can't do the same thing when English is their first language...I guess it's because they're not paid to ask intelligent questions, just to entertain and polarize people
+danimalplanimal Loving what you do sure helps a lot. And you can't buy that with money ;-)
+Nikola Danaylov Respects from my keyboard as well.
+James Nation True, that is how interviews should be conducted but sometimes you have biased interviewers with an agenda who would always try to direct the interview according to their agenda.
This guy is doing a good Job. Well done!
Thanks to both of you for this interview. Clear thinking is the rarest most precious ability one can have.
Chomsky isn't a neuroscientist.
Chomsky isn't a programmer
Chomsky isn't an AI expert
Why is his opinion relevant.
Cronos Perhaps because his 60 years of presumably very intelligent research and reading can be assumed to have crossed into and covered those fields to an extent that makes his thoughts on those subjects very worth listening to.
I would not be too quick to pidgeon-hole him or people in general. Besides, he is quite possibly a reasonably proficient programmer. I started programming in '83, and most programmers are reasonably proficient within 2 or 3 years in any one language, especially if they have the basic ability to think that way in the first place. I would guess that Chomsky has always had that basic ability, and after 60 years at MIT, has had quite a bit of exposure to programming many times in that time, especially given his field of study: linguistics.
David Tangye Yes but general programming and artificial intelligence programming are completely different. AI techniques have been evolving and changing at a steady pace throughout the last several decades, and the biggest changes are happening recently.
Cronos The guy has degrees in the human mind, language and Cognitive science which is why his opinion is relevant. He was the one that defined language and speech properly for the first time in human history (if i am not mistaken) or the parameters to considers what makes it real language human from other animals such as gorillas or programs etc. It is pretty interesting stuff.
csmain I'm into linguistics so I know about a lot of that. Once again, I don't see how that makes him as qualified as someone actively working in the AI field.
+Cronos I agree. Despite Mr. Chomsky made some very good/relevant statements, but to qualify Watson as a PR gag shows that he does not fully understand Watson's achievement and it's future potententials.
"that's fine." must be the 1960's version of "that's cute".
I wish Noam Chomsky was my grandpa.
Not me :(
"Grandad, can I have a candy?"
"Uhhhhhh, did you know that because of Western foreign policy, you're part of 1% of children in the world able to afford candy? Some children have no concept of candy. To me, there's an imbalance between your expectations of candy and your right to candy, that needs to be addressed."
"Forget it."
@@harmonyjones8035I bet his grandchildren got a candy. :)
One of the most decent human beings to walk the earth.
I would sit all day with him getting as much wisdom as possible.
I had college class about 18 years ago where Noam Chomsky was required reading....some his profound commentary into our world and environment were so true .....i still quote him today with all kinds of folks and am amazed at how many people say "Chomsky who??" His work should be required reading at late elementary school or middle school ......when the truly young can absorb such wisdom about how we think and how we want to grow up and effectively shape a better world where a visitor from mars wouldnt have to come here and observe our insanity as a species!!
Don't think Chomsky is a 'sociologist' -- he understands the science of intelligence from a lot of angles. If he says that computer programs are nowhere close to real intelligence, and that the impressive AI progress we see is from the 'brute force' application of low-level processes, it's worth considering what he's saying. He's saying that Siri, Watson, Deep Blue, etc. may seem smart, but it's imitating intelligence, not manifesting it.
But " imitation is the sincerest form of flattery". Please forgive my appropriated 'Colton' quote. I do agree with Chomsky.
+ambassadorportal
Are you sure he's saying computer programs are nowhere close to "real intelligence"? My perspective is that brute force or otherwise, if we're talking about phenomenon that can be modeled in terms of functions (and I can't see why that doesn't include everything we consider indications of "intelligence") then whether or not it might be considered imitation is irrelevant. If it quacks like a duck, looks like a duck, .....
This might be intellectually unsatisfying to Chomsky, you and even me, but having a compositional understanding of the functions we approximate with machines is *not* a prerequisite to modelling them to a sufficiently indiscernible approximation to what we consider is the "real" thing.
In other words, I think it's highly likely we arrive at software machines that produce signs of intelligence indiscernible from our own, long before we "understand" such systems. I don't think complex systems need to be understood at all. We only need understand how to construct machines that can can mimic complex systems as approximated functions.
@@diwr I like what you're getting at - in terms of the 'minimal ontology' when it comes to ai construction. problem is, chomsky isn't focused on the performativity of the engineering of the systems which he says is "fine" a la "brute force," in the same sense that a Ford Mustang GT that today puts out 570 HP (as compared to the 300 HP in the old days) is all well and fine. his point is that there's no reason to entertain wild fantasies (a la kurzweil) about a future w/ machines as our overlords. (as an aside his point that we should focus on the actual overlords is pretty on point). In his rationalist philosophical view, to understand real intelligence we have to understand the nature of choice, freedom, creativity. we're a long ways away from the day when the computer will walk into the bosses office and say "hey boss, i've got an idea"
i wonder what he thinks of alphazero and alphastar.
Very good interviewer, doing what so many others fail to do - keeping his questions brief and clear. Great interview.
Thank you very much for producing and posting this interview.
Many users have stated that they respect Chomsky but because he doesn't think much of the Singularity he is "not in touch with science and technology today", "completely stupid", an "idiot". Just because he disagrees on this point he is invalidated and vilified. Shut him up, he doesn’t agree with us! The ad hominem attacks are quite shameful and reveal the real driver behind the friendly mask here: fanaticism. People here are defending the Singularity with religious fanaticism and illogical and emotional responses. People who have different opinions are not allowed, people who won't toe the line aren't welcome here. Some posters are just as closed-minded and brainwashed as racist bigots, materialists and religious fundamentalists. He disagrees and therefore he must be cut down and discredited. It's very disappointing.
"I don't particularly" LOL..
This is the human mind - the genius mind of subtle , deep awake understanding. notice the reluctance to speculate. Noam preserves the sacrosanc kernel of human consciousness and yet displays it for us in this interview .. thanks Nikola.
This is one of the most significant interviews I have viewed on the new information path, which has become known as the internet. Thank you.
I feel very lucky to have found this interview!!! Really good an memorable interview of Noam Chomsky about a really important topic! Thank you very much! I just subscribed to this channel.
Good interview. I could listen to Chomsky for hours.
I enjoyed this interview. Very insightful and the questions were fine!
Honestly, asking him about zeitgeist movement? Might as well ask him whether he thinks horoscopes may be a good way to decide foreign policy, or how to talk to women.
Ƹ̵̡Ӝ̵̨Ʒ that is, if one buys into your ignorance
***** I think the appeal of conspiracy theories is the same as the appeal of religion. Someone is in charge, power concentrated in a single place, and you know about it. Everything happening is part of a greater plan. You know the ultimate truth of the world.
Did you actually check the claims against reliable sources? You'll find that most of it's just made up. I'm also a hobby hypnotist and I recognize the techniques when I see them. The zeitgeist video is terribly unethical. Don't trust someone who didn't bother to respect your mind.
Ƹ̵̡Ӝ̵̨Ʒ The Zeitgeist Movement is actually just a repackaging of Communism being promoted by modern Socialists. It is merely a clever smokescreen used to recruit new blood to the cult of Marx. Some people just refuse to learn, so History repeats itself.
HeartlandRanchTV I find the Nordic model of social democracy with NZ-style proportional representation and approval voting most reasonable.
pretty sure none of you ever read Marx... in the context of his time... so, whatever
*Great Interviewer. He was not prone to interrupting the interviewee like most others. Great Job!* Upvoted.
Do not forget that Chomsky is an expert in lenguages and language is the most complex thing that the human brain is capable of doing. He understands, through his knowlege, that thought is something much more complex than simple Intelligence.
Do not forget that you are in no position to declare where the human brain does (or will) peak unless you both, lead the field in all areas of human intellectual endeavor, and have somehow managed to gain perfect knowledge about all the human mind ever will accomplish.
I wouldn't argue that Noam is unintelligent, or that he's not entitled to his opinion. I do, however take issue with the idea that language is "the most" complex thing a human brain is capable of, and the implication that studying it somehow puts him among the greatest minds on Earth and giving him insight into what other fields of study may accomplish. Seems to me that's what the comment I was responding to attempted to say, and it's bullshit.
In fact... just to back pedal a little bit, MOST people whose profession involves specialization are bright enough not to comment on the possibilities in other fields to avoid getting egg on their face. Just try to get a chemical engineer to design a structure for you, or have a lawyer make a pronunciation about the safety of any structure that might result. It takes a lot of short shortsightedness, arrogance, or stupidity to boldly make predictions as to what will never happen in specializations alien to your own.
Zaphod7835 The "deep" of your comments makes me realise that you're less qualified than Chomsky looks to be about the singularity topic . The complexity of natural language is a fact, but if you don't know this, sorry, blame your ignorance. So, is clear where the bullshit comes from.
Excellent interview. It helped to clear up many of my misconceptions about A.I. and the Singularity.
Given the questions and the short time, his answers were straight forward and interesting to me. I enjoyed the video. With my recent interest and understanding of Singularity, I'm looking forward to watching more videos overall and understanding different opinions on the matter. Thank you for the video.
Chomsky brings reality to the discussion, and that of course isn't wanted in the overly optimistic future vision, that enables people to drop activism on real issues and wait for the robotic messiah prophesied by Kurzweil.
'peasants, craftsman, labor'
'rent Democracy'
'the only human property is greed'
'mode of social organization that reflects other values'
the evolution of Noam 'the Imposter' Chomsky'
Noam Chomsky IGNORES this principle ~
OBJECTIVE CONTRADICTING HISTORICAL FACTS and mountains
of empirical evidence that his words and ideas simply IGNORE
his analysis of US and Western civilization is EXACTLY
verbatim what and how Stalin, Mao and even Hitler said.
He believes he is an 'avenging angel' against the evils of mankind in the capitalist
free markets World we live in. This is what super egotistical sociopaths believe.
Even his robot like monotone speech pathology is a dead (sic) giveaway
for his personality disorder, his grandiose thinking and reasoning process
is pure nihilism really SUPER self centered anarchy. A dream whereby
the State with a
'mode of social organization that reflects other values'
'other values' ~redacted, to be determined by ?????????????????
I have listening to Chomsky for decades.
I could bury Chomsky with over 40 years of
experience of facts,bibliography, notes, sources,
reading, debating against the self righteous tripe Chomsky espouses.
In fact, you will NEVER hear or see Noam Chomsky in ANY serious
one on one extended, unedited debate. He would simply leave the dais.
He never REACTS to FACTS that refute his arguments.
Chomsky goes into a 'fugue state' and repeats himself literally
continuously, especially in his 'books.'
Psychologically
its sad and scary that so many still revere or listen to him.
There are so many better academics that could argue his
basic talking points. His admirers are all in a dis-associative denial
Stockholm Syndrome fugue state. Like the bogus 'capitalist' PRC with
a portrait of their very own mass murderer, Mao, on their currency.
I began my education at 11 years old watching
Gore Vidal debate William Buckley. I know exactly how to destroy
ANY Marxist, deconstructionist, nihilistic, Orwellian, socialist
apologist or revisionist.
My #1 argument and evidence against Chomsky and his facts ?
DEMOCIDE, specifically by left wing governments that used
the exact same examples and words of Noam Chomsky.
120 Million human beings murdered in peacetime in one Century
by their own fellow citizens and countrymen because of what people
like Noam Chomsky say and do.
So, pardon me if I have offended your sensibilities and Chomskys
'apologies' for the Left Wing's solutions in the past of
stupifyingly simple mass murder for us that don't get Noam Chomsky.
@Christopher Sample
Noam Chomky''s................last century......
Karl Marx's idea's (socialism) murdered at least 120 million human beings.
ethnic Russians and Chinese by forced labor, starvation and disease.
DEMOCIDE, a 200 year old French revolution word meaning
murder of your own Citizens in peacetime.
Your 'happy families' and lives under Marxist economies
is total BS.
BS because all the "problems" in Marxist societies we're (are)
murdered. The people who survived and thrived under
any Marxist country we're simply lucky to be alive
and reaping the benefits of MASS MURDER (of the "problems")
never seen before in all of history. Stalin loved murdering enemies of "Socialism."
So, did Mao Tse Tung in the PRC.
Cultural Marxism was Leon Trotsky's idea which he was murdered
by Stalin for. Trotsky and millions of followers who kicked off their
100 year old "Cultural Marxist" revolution in Frankfurt, Germany in 1923.
"Critical Theory" is actually a theory Socrates taught
and which cost him his life. Its the reason Athens became a
ghost town within 20 years of his death. It simply means
tear down existing public and private institutions with relentless
logical fallacies.
Cultural Marx survives in the every Western and Eastern institution
where Liberal Democracies have flourished world wide by
CAPITALISM that feeds 2 billion people a year from starvation and lifted
1 billion Chinese and ethnic Indian nationals into the Middle Class
by trade with Western Capitalist nations in the last 50 years at least.
@Christopher Sample
Its all in the Manifesto of the
Frankfurt school of "Critical Theory"
its called Deconstruction. Actually. since
Socrates via Plato and Revisionist History.
History from the point of the losers, who we're defeated. Defeated in the war of ideas. Leon Trotsky was murdered by Stalin
for these very reasons. Marx was an intellectual sadist. Pain was his answer to a reticent bourgeoisie belief system.
He used Hegel's 'dialectic of reason' to convey his #1 logical fallacy in the
'dialectic of Materialism' and a zero sum game of self destruction as a basis for
a Cultural revolution that Orwell and Huxley saw through.
And today, exemplified by the likes of the
oldest Democratic body in the World, the
800 year old British Parliament
versus BREXIT and the EU and UN and the Socialist NWO of a SUPER Government
(unelected, EU Commission) under a false flag of egalitarianism and 'Democratic' social engineering, relentlessly repeated by people like Noam Chomsky with Marx as their prophet of Government by semantics, pain and debt slavery mixed with Freud's prediction of a mass Stockholm Syndrome and Narcissism, especially financial narcissism. Whereby,
'Repetition of Disinformation' is a KEY tool of control and indoctrination.
See, for a definitive over view of my statements/views in
"The Culture of Narcissism," C. Lasch 1979
@Christopher Sample
WELL your comment.........................
"Chomsky is the smartest and most moral person I have ever listened to......................"....and.....He speaks truth , not to power, but to his fellow human beings "
that's what MY post was ADDRESSING.
Chomsky is a man with a single pHd in linguistics who is probably
a classic sociopath with nihilistic tendencies. His mask is a dead prophet
of the Dialectic of Materialism Marx. Especially, Marxist semantics.
PS lots of BIG IDEAS to 'think about it'
@Christopher Sample I doubt very much you are UN intelligent.
Chomsky is a very intelligent man as well.
My strident comments are just that VERY strident. And, yes I have been
approached by 'Think Tank(s)' for thinking out of the box which
most business's and Governments 'war game' continuously.
Hats off to the interviewer! Polite host. I like polite hosts.
"The question we're trying to ask is: 'Can we define a theory of being smart?', and we are eons away from doing that." 💯💯
A very good, refreshing interview. It's always great to hear Noam's thoughts on modern science and philosophy
Wow Velis, the Feynman clip is an absolutely fascinating classic video - I will re-post it on Singularity Weblog. Thank you very much!!!
And now we are here.
not much has changed.
I disagree@@fletcher9328
That was fascinating.
Thank you very much Sage Mantis! I am very happy you enjoyed the interview so much!!! ;-)
One of the most insightful interviews I've see with Noam Chomsky lately (while talking about something else than politics). That's where we can see how briliant he is. Bravo Nikola!
Chomsky appears unaware of the breakthrough in neural network algorithms ca. 2009, chiefly due to Professor Geoffrey Hinton, and now known as Deep Learning. He seems similarly unaware of the breakthroughs in natural language processing and image processing, whereby Deep Learning has blown away all the older, established paradigms. It is uncomfortable for me to listen to an obviously intelligent individual like Chomsky speaking from a position of ignorance with such confidence.
+Andrew Palfreyman - Just curious - what has these breakthroughs in Deep Learning produced that has invalidated Dr. Chomsky's positions?
+Andrew Palfreyman Deep Learning did not really "blow away" older, established paradigms. It can be considered a (more) efficient approach to learning... like a better algorithm. More will come. But indeed, Dr. Chomsky seemed to reject that such small useful steps could lead to singularity.
how did u get 30 min with Chomsky ?
He seems to be one of few public figures, explaining how limited ( and useless ) the current field of AI is. Scientist are struggling to control even the simplest insects, and there is no self-autonomous and self generating computer program, as that is ( correctly ) not how programs are written, and even if such a program was successfully written ? it would still be limited and physically constrained in some form of RAM memory. I.e. You could irreversibly kill it, just by unplugging the electricity. And then, you would have to start over, and, you would get another result, i.e. another ( not quite identical ) AI program. Also. U will find these fractal-like programs, fill up thousands of gigs in seconds, so they need "vents" ( channel "pre-sets" ) where to dump excessive information, no one has constructed such vents, that doesn't immediately kill, or render the AI program useless.
Driver less cars are a fad. They will only be functional on separate roads ( droid tunnels ) even "if/when" they can prove 99.99% security. Likewise, with the flying drones of Amazon. They need separate air channels, that only the Government can free up. It's all at a "cave man" stage.
***** I don't think I wrote anything about predicting "the future", whatever "predicting the future" means to you. To me, it's meaningless, because any machine ( or organism ) that could predict the future, would automatically create the future, and then it would no longer be a "prediction", but more of an unstable feedback loop into that "particular machine", which would be impossible to stabilize, by any outside influence.
There is no current "field of AI" on planet Earth, beyond programming, which means machine coding BY HUMANS. It's also a human coding, which is highly imperfect and mixed into social/economical status, in which a lot of code is being overtaken ( made abundant and redundant ) by "other" groups of humans, instead of anything remotely close to automation, by having software which can truly create software ( and itself ) Instead, we have macro tools, which can be very "cool" and time saving, albeit dead. The field of AI has a total progression, of exactly zero, and the evolution of software is currently a 100% race between humans, according to various market principles, mixed in with a lot of government subsidies. There is a struggle, but it's an economical and human struggle. Current machines don't give a fuck. So, until you can create a smart car, that actually wants to drive ( that is excited, and happy to drive ) you will create nothing more than cans, running down a track, according to buffer rules, or decentralized network ( crowd ) theory, if sufficiently advanced. But, if you are truly interested in AI ? you will study animals, humans, and even music or economy, before anything purely technical, because the entire field of science and technology is at total zero ground, in comparison to true AI.
I have a solution. Replace the word "AI" with "coding", and all your thoughts and discussions, will be technically correct.
( I don't know the future, but I know the present when it comes to this question )
3waybar "To me, it's meaningless, because any machine ( or organism ) that could predict the future, would automatically create the future, and then it would no longer be a "prediction", but more of an unstable feedback loop into that "particular machine", which would be impossible to stabilize, by any outside influence."
The brain evolved as a mechanism to predict and plan for the future state of the environment around the creature that's in it.
We predict the future, then we either sit still and observe it, or we act to change it. We are in essence, a prediction engine.
So many confuses the issue of developing AI with the skills lacking in individual humans. Sure, no single human will ever create a fully functional AI. And a single human being, will never create a CPU from scratch. It takes an evolving effort that builds on the previous generation. And that is how we will build AI, by making algorithms that builds upon itself and improves.
The AI scientists and engineers are studying all aspects of animal and human behavior, to understand before we can make. Examples of cars having intentions and wishes are silly, at least with today's knowledge of what might happen and our needs. Obviously the car will have to have some sort of reasoning in the future, if there is imminent death of people it must make a decision on who to save and who to kill.
Full brain emulation won't arrive until after we have AGI, a lot of the processing in the brain can actually be achieve with simpler algorithms and mechanisms, compared to the biological brain.
***** "Wind mills" are already here too, but in my country 99% of the mills are subsidized by the Government, both in construction and upkeep. What pays for that subsidy is ( World record high ) wage and consumption tax. I.m.o. not a very "smart system", and the issue with "drone cars" is also that Government will probably build the special channels ( tunnels ) for this kind of automated traffic. It's at risk of becoming yet another cluster-fuck. Not so agile and "smart" ,as it can appear on the surface, and certainly nowhere near, even a simulated AI.
These "drone roads" will have strict rules ( much stricter than current human driven traffic ) because "the road" and wholesale movements is the challenge, not the singular ( alone ) car. It doesn't matter if Google's market cap is 300 B or even 2000 Billions. In the end, the entire "smart car" market will be a 100% Government controlled, because the only smart thing that needs development are wholesale movements of traffic, and smart roads. I.e. the intelligence needs to be build into both the road, and the surrounding environment, like winds, ice formation, etc.. to no end, and, since humanity does not possess true self-programming AI, the Government will rule everything, "top to bottom", which will probably increase risk, no matter how much computational power you can store, in each and every "car". The "drone car system" will also become so effective, that it will no longer be anywhere near the current configuration of personal ( or singular ) vehicles, but more like packets of information, like the "bits" on today's Internet.
The self driving car, has been a reality for decades. What we don't have is a fair, simple and just "societal system", to implement it, and everything that follows. We don't even have any valid or tested theories, of what is to come after capitalism, when owning a car will become meaningless. We don't have true AI, on any level. Not in computers, not in law practice, not in Government. We are at zero.
Sondre Bjellås To me you are simply saying, and I also thought this during the interview, science fiction is only fiction until we read about it in the papers. Plenty of scientific advances have been written about by authors ahead of their time who usually had some basic understanding of the field or even similar philosophical pursuits as scientists who gave them the ideas in the first place.
Humans talk about their ideas and look to understand the universe, for the simple reason to share these ideas and understand more and more as you say, thus building upon human advancement no matter the field/application. Language is modus operandi of the human race for almost all information (of import or no), and so a strong understanding of it can give you insight into the motivations of not just individuals, but whole civilizations and their histories. I daresay no man who has lived has understood these concepts and the implications of linguistics quite like Noam Chomsky.
Wonderful interview. You are very lucky to have had that opportunity to interview him.
Thank you so much for this. Noam was my virtual mentor in Communications at Condordia U in the '90's. He shaped the vision of my media watchdog philosophy. He helped me to be a critical thinker. thank you for these questions.
I advise against reading the other comments... It's a bleak landscape.
Every several months I come back to this interview, because Noam is literally the archetype of Dumbledore in terms of a human being. He makes me feel like even though I think the existential nihilistic perspective is correct, and I am a mortal and anything can happen, and that I don't know where things are supposed to go, that everything is alright because he's kind of seen it all and he's completely aware, wise, 'there', and calm. I think more than anything, he's the universe's treasure for that reason on top of the genius and resourcefulness. If he can be okay, so can I. Assuming I'm not suffering or something. And that's where Aubrey de Grey and Jacque Fresco come in. Hopefully we can live long and prosper until death is synonymous to simply fulfilling your focus and that's it. The Venus Project was recently awarded by an organization supported by the United Nations and has after 100 years of living and 85 years of working on his project successfully transitioned to the next phase-- they are now constructing their first experimental city. I cannot find information beyond that, but perhaps, Nikola, if you could possibly, potentially get ahold of Roxeanne Meadows or Jacque Fresco again, but it is likely he is beyond the point of interviewing. It's a pretty rare occasion that someone so important also happens to be among the tiny fraction who lives to over 100.
Still come back now that we are post singularity?
Downvoted for the clickbait title. Chomsky gets to his bit about Ray Kurzweil around 15:20.
***** yeah, but only like a 3 minute segment of the interview.
BTW - I would have liked to see you press him on that point - isn't DNA like a program? But I can see how talking to Noam would be intimidating and you didn't want to get to contentious especially at the expense of avoiding other topics, etc.
+Benjamin Quellier Thanks for summarising the talk.
Don't down vote the dude, Chomsky's been doing clickbait since clickbait existed, it's not the channel's fault.
Seriously, Chomsky will be like "The United States is the largest supporter of terrorism in the world".
And people would be like :O "What do you mean?"
"The American drone program is the largest state sponsorship of terrorism in the world, there is no other state program that kills as many people in as many nations without a declaration of war."
Seriously, that's how Chomsky frames most his points. I mean, he usually does it with much more grace and detail than I just did, but that's the basic idea. lol
Dr. Noam Chomsky is a famed linguist, political activist, prolific author and recognized public speaker, who has spent the last 60 years living a double life -- one as a political activist and another as a linguist. His activism allegedly made him the US government's public enemy number one. As a linguist he is often credited for dethroning behaviorism and becoming the "father of modern linguistics" (and/or cognitive science). Put together his accomplishments are the reasons why he is often listed as one of the most important intellectuals of the 20th century. And so I was very much looking forward to interviewing him on Singularity 1 on 1.
Unfortunately our time together was delayed, then rushed and a bit shorter than anticipated. So I was pretty nervous throughout and messed up some of my questions and timing. Never-the-less, I believe that we still had a worthy conversation with Dr. Chomsky and I appreciate the generous though limited time that he was able to grant me.
During our 28 minute conversation with Noam Chomsky we cover a variety of interesting topics such as: the balance between his academic and his political life; artificial intelligence and reverse engineering the human brain; why in his view both Deep Blue and Watson are little more than PR; the slow but substantial progress of our civilization; the technological singularity...
Category
Science & Technology
License
Standard UA-cam License
Thank you for the upload!!
This is a precious video.
Everyone hiding in holes, plugged up to a computer. Pale, lonely, mangled, enslaved.
You make it sound like a bad thing
I studied anthropology so I like Chomsky. I also don't believe in the quasi-religious notions some have towards a technological singularity.
However I don't think he argued his position very well at all.
+Hoyt Mueller what position? on whether or not star trek is real?
+Hoyt Mueller Yea I feel like he too easily dismisses stepping stone tech. Its really bizarre. Even if he doesn't think we are making headway on solving AI problems, he does a terrible job of explaining why he thinks that.
It makes me qualified to like the guy because I've studied his work.
I also said he didn't argue his position well.
My argument wasn't against AI, just the "quasi-religious notions some have towards a technological singularity."
Most people that discuss the singularity think of it in terms of an age of magic where literally nothing can be known about it.
It is my understanding that even if the processes behind something cannot easily be explained, people can understand what technology that is "seemingly magic" does and what it means. Someone 100 years ago would understand the implications of a smart phone if explained. They may not use it that well, but they'd be able to use it.
Same goes for the singularity.
+Hoyt Mueller I disagree. His point is that the complexity of the processes that give rise to human cognition are immensely complex, are still poorly understood, and will likely remain poorly understood well into the future. The "singularity" and like ideas, for whatever reasons, do not take this sufficiently into account.
Chomsky put that across quite clearly to me. If he were to attempt to take the explanation much deeper than that, it would quickly get bogged down in very technical gobedleygook. So, I say his point was as clearly stated and well argued as can be hoped for in the course of a half-hour conversation targeting an audience of non-specialists.
+Hoyt Mueller
How would you have argued his position?
Great interview. This is the first time in 33 years I have heard or read Chomsky and liked him.
Most provocative and piercing interview. Unlike so much on the net today, your interviews are well constructed and enlightening. Kudos.
On the "miracle" of avoiding nuclear war. I think Frank Zappa said it best.
Paraphrasing, "... considering the multinational corporate machines that make global political decisions, nuclear war would be really bad for business."
+xoco nostle rip Frank Zappa
"Aerodynamically, the bumble bee shouldn't be able to fly, but the bumble bee doesn't know it so it goes on flying away." (Mary Kay Ash)
Actually, they figured that one out.
Thank you very much friend, that was also one of my favorite quotes ;-)
Nikola - I absolutely LUV that you present Both sides of arguments...... you are the man my good brotha! Keep up the good work my friend.....
+james thomas Thanks very much friend, and you bet I will ;-)
+Nikola Danaylov Subbed bro
The final question- thanks for framing it completely open ended. Of course, the answer would be much broader in scope than AI..
“…now any rational person recognizes this [global warming] to be extremely serious, and unfortunately any rational person doesn’t happen to include any large part of the U.S. Congress.” Noam Chomsky
Excellent questions and dialog. Thank you for sharing.
Nikola, wonderful job bringing Noam into the discussion. As a novelist I need to look at futuristic issues with a fish-eye lens, bringing into various points of view so as to transmit them to the reader through my story telling.
.
Noam Chompsky is a brilliant man that has had an extraordinary career. I have great respect for him. However, I think that he does not appreciate the power of exponential growth. Ray Kurzweil makes a great case for the Singularity in his book. While it's easy to dismiss Ray because his predictions seem outlandish, I still haven't heard a reasonable argument against him.
His predictions aren't simply outlandish! Majority of Ray's predictions of things we would have like 3 years ago never came true at that time and haven't still come true years later. He is a man who gets too much attention and support for spewing bullshit. As the Bible says, you shall know by the fruits of the tree...(not the empty words everyone can shout!)
TheVariableConstant Would you be so kind to provide some examples of the things that did not come true? Thanks.
+Dalibor Tesanovic the video I watched was many years ago. I don't have time to look for it. An example of something he said was glasses that project the display onto the retina would be common. He said we would have that by 2010 or 2015, I don't remember. Even now years later I don't think there is such a thing in prototype let alone being common. His fanbois use google glass as a cop out explanation. Google glass is no where near the same as projection onto the retina and it was even a failed project that didn't become common.
TheVariableConstant
2019.
And yes, viable virtual retinal displays exist. Full augmented reality is getting closer as well. I've heard of worse predictions of future technology.
Yeah, that's pretty accurate in the prediction game as far as I can see. You have to look at the root of what he was predicting too...wearable tech that was directly in front of your eyes and provided information. In that regard he was spot on, pretty much.
Even so, Science Fiction leads the way. There is a certain mysticism at work whenever someone just insists that the Human Mind can't be reverse engineered... whatever we might want to say, in actuality the brain is the seat of consciousness, and simply matter.
Like everything else, Science will unravel these mysteries, and Science Fiction is a portent of this. So sure, let "The Singularity" be Science Fiction. This doesn't make it any less of a powerful idea.
How many other things that exist right now were just science fiction? Some of them, like self-driving cars were just science fiction 20 years ago!
That said, I'm not a fan of obsessing over the singularity.
"...the brain is the seat of consciousness, and simply matter." < Maybe.
lol.
please.
Tad Donaghe: _"Some of them, like self-driving cars were just science fiction 20 years ago!"_
__
en.wikipedia.org/wiki/Autonomous_car
__
en.wikipedia.org/wiki/EUREKA_Prometheus_Project
Thank you for your well informed opinion on a subject you clearly know so much about Mr.Chomsky, a great representation of your technological knowledge and your overall scope and vision of technology. You clearly know what your talking about.......
The subtitle of this video should specify that the singularity in question is the technological, as opposed to the cosmological singularity.
Full marks to the interviewer - excellent questions, and Chomsky himself, at 85, remains the most intelligent and ethical human I have ever (indirectly) encountered. He hits the nail on the head every time, it seems.
Aged like milk :) I imagine it's very hard for him to realize that he was wrong all his life
Thank you very much for the interview. It is amazing how many different brilliant people US have, yet their government is total degeneration of humanity.
Ich bin begeistert von diesem Professor und wünschte er wäre jetzt Anfang zwanzig, um noch lange wirken zu können. Der Ton machte es etwas schwierig alles zu verstehen. Die Fragen waren sehr klug gestellt. (Im Fernsehen ist das selten der Fall.) Ich wünschte, dass jemand wie Herr Chomsky Zeit hätte an Schulen zu gehen um mit den Schülern zu diskutieren.
Bravo! Thank you for this! Gave me some unusually deep insights -- and actually hope. Thank you so much!
Hello. if you're interested in artificial intelligence, you can check out my ebook “HOW TO CREATE MACHINE SUPERINTELLIGENCE”. This e-book is available for FREE in Amazon Kindle Store until 2 April 2017.
It includes the following topics:
Intelligence as information-processing system
Classical and quantum computations
Artificial Intelligence and Machine learning
Artificial Neural networks
"Its kind of a miracle that we have escaped a nuclear war. Literally a miracle and you cant expect that miracle to continue, so something quickly and urgently has to be done to eliminate this curse."
This is my favorite Chomsky interview ever. Wonderful job! I've never before heard this brilliant and wonderful guy ever be anything but a severe critic! Here we finally glimpse his positive side, and find out what he himself believes should be done. On top of that, we find him knowledgeable in funny ways outside his field. I think he even smiled... and not while leveling some sardonic wit! I'm impressed! Chomsky as an optimist?!? I may need to be resuscitated! But it's so cool to see! You really got something from him.
Underneath his cynicism, there actually is quite a bit of hope. It may be hard to find, because he spends most of his time addressing the problems of the day, but it's there. He does make it a point in some lectures to remind us that the problems we face aren't new, and that the general trend is toward improvement.
xyZenTV
Thanks... Yeah, I always felt it had to be there, or there's no way he could have been so entertaining all these years.
I feel like, totally comfortable in this old school stuff. Even though the day seems headed for the final curtain. The whole scene is a museum piece. I really want to meet some of these greats before it's too late. Marvin Minsky is my personal fav.
*****
I Totally agree. I have the idea that he exemplifies the 2nd thread (I count 4) of human discovery that is natural science. MISS IT! (I was educated in Theoretical Biology.)
stancurtin he stole my words that's okay &
A true intellect, clear logical reasoning without sensational claims for publicity.
Great intellectual living in our era, very insightful discussion.....
Singularity isn't about machines being able to think independently. It's about adding technology to the human brain until it's dependent on it. It's about Techno-human hybrid that will gradually become more techno than biological.
"We are not angels, we are organic creatures like all others."
I don't see any elephant creating autonomous machines. But we have. We have achieved what was once science fiction.
And we will continue to create exponential growth in the future based on the technologies we are creating today.
What exponential growth? We ve used up nearly all of the natural resources that we need to power growth and the planet is dying...
It was very generous of Chomsky to give this interview.
Hey!My question is do we should hybrid thinking now or at the end(when we have nanobots in brain --we connect the cloud) or at the big computer--singularity times ? ... thx
The singularity cannot self assemble, we have to build it.
But we do not comprehend the mind to start with.
If the brain was so simple that we could understand it, we would be so simple we couldn't understand it.
what do you mean by "we could understand it"? do any of the people watching this understand how the personal computer operates? yeah, there might be some people who understand certain aspects of PC architecture, maybe even in fine detail (extremely rare), but all of the PC, all of the physics, chemistry, technology, logics, programs behind it well enought to be able to replicate it? I doubt there's even one such person, anywhere, in the entire World. yet computers are commonplace. So collectively we can understand something too complicated for individual to even begin comprehending. On the other hand, AI doesn't have to be based on the model of our brain. It is certainly not the optimal tool for thinking (as our history suggests), - there simply must be better ways to reason than by using the copy of human brain. For example, digital cameras, sensors are not based on the model of our eyes, - our eyes are not the best tool for seing things, they have no zoom, suck at night, and peripheral resolution is terrible ;)
@@k.squared The thing is that we don't have any understanding of how the brain's actual reasoning capability works. Yes, we might be correct at how neurons work on an electrochemical level but we have no real idea how they group together and what causes them to adapt and learn the way they do. Not only that but we don't have any kind of working computer model for reasoning capability. Right now the AI we use effectively cheats. It essentially just runs millions of simulations. It doesn't actually reason about anything. It's the equivalent of giving it a highly complicated maze and letting it find a way to solve it. It has no real conceptual learning.
@@taragnor Effectively cheats? how the hell do you think we learned what we now know? How you can't comprehand that the learning methods used in ML are the exact copy of what we have been using to evolve our own species is beyond me. And yes we do have a working computer model for reasoning capability, we've had it for a couple of years now but it seems you've been sleeping under a rock for the last decade.
Thats the point of AI. Understanding why it does what it does is the last stage of development and likely one that will never be achieved. Its the
equivalent of an ant trying to understand the politics behind WW2. Its not realistically possible.
@@Wowplajer : No. It's not an "Exact copy" of anything. That's one of those misconceptions of AI. Artificial Neural network isn't designed based off some advanced neural discoveries and modelled after the brain. It's just a clever buzzword. The original concept was designed was back in the 1950s. It's not like some artificial brain. It's essentially a mathematical trick. It's an algorithm like any other computer program.
As far as real reasoning capability, I've never actually seen it in an AI. The very nature of which they're trained, doing millions and millions of simulations, basically implies no reasoning capability. I mean, when you learned how to drive a car, did you have to crash it virtually millions of times before you figured out how driving works? No, a human doesn't because you can reason out what threats you have to avoid. For an AI, it really has no idea. It effectively flails about at random until it finds something that works. Seriously just watch the iterations of an AI learning and you'll see it's really not smart at all.
I think Chomsky's position is right. Singularity is fantasy in its current form.
*****
I admire the way you handled the interview; and drew Chomsky out on many important issues. His wisdom is profound.
I disagree with many posters here who say Chomsky doesnt understand AI. In fact it is they dont understand the fundamental issues he raised.
Chomsky has been studying the philosophy of mind and has been engaging with the subject of AI at MIT since the 50s but the random people in this comment section say he doesn't know what he's talking abut so I believe them.
I have no idea how you managed to nab so many great interviews, lol. Awesome work.
I don't see commenters clearly state the central argument of inevitability of artificial intelligence, so here goes: 1) Either thinking is completely physical, or there is a supernatural element, i.e., a soul. 2) If it is purely physical, then the brain is a thinking machine. 3) We know enough about the brain to know that it uses both electrical impulses and physical movement of molecules to operate. 4) Any analog architecture of the brain can be simulated to the desired degree by digital processes. In fact, we know the universe is digital at the quantum level. 5) Computers are much faster than the brain because there is much less physical movement in digital circuits, which are operating at the nano scales. Just compare the limits of human visual systems with the tracking cameras in baseball, for example.
Suppose that we are able to understand the brain's architecture. All we have to do is copy that architecture into a computer. Even if the AI is only as smart as a human it would operate at least 100-1,000 times faster, be able to design and create any body parts it desired, have virtually unlimited memory, and be able to clone itself.
Now imagine such an intelligence trying to solve a problem like how to build a better computer. It could clone itself up to the limit of hardware space, and for every year of human time they are working 300-3,000 years on the same problem.
There are only two ways that the singularity could not happen: 1) thought is a product of supernatural occurrences, or 2) we can't figure out how the brain works.
Chomsky claims the second. However, look at Deep Blue and Watson. It seems unlikely.
I am beginning to think consciousness is just the brain though. If we don't understand the brain, why are neural implants and computers that turn brain impulses into commands taking off?
Kirie Sakurame yeah there is not really any need to understand the brain to create intelligent problem solving machines.
We probably don´t even need to understand the brain to build a true AI. Learning programmes could achieve sapience over time. Think about it. A self improving program, which we actually can to some degree build already and use, operating in the internet. That is the likeliest candidate to achieve sapience. Getting smarter and smarter, learning and evolving. I believe that the first AI to achieve that will not be created, but just like us suddenly start to ask THE questions and so to speak, come down from it´s metaphorical tree.
A I as we know it doesn’t try to copy the architecture of the mind, so it is on the wrong track. It uses algorithms or brute force, not neural networks interconnecting the entire knowledge space. It is confined to isolated tasks, such as driving a car, translating languages, playing a single game.
In 19:19 "...technological singularity is science fiction..."
Hahahahaha have fun in life not understanding that this is going to happen lol
i learned chomsky grammars in university. i got best note on that topic :D and we wrote programs based on his work to check grammar and thesaurus in letters.. compiler need it to translate programming languages
Hi Nikola as a proponent of the singularity, great of you to also upload others perspective. I don't believe in the singularity but the book The Singularity is Near is still one of the coolest books.
The Age of Spiritual Machines took me like a storm when I read the Danish translation years ago. Now I'm skeptical to say the least
+Victor Alexander I should say that I do believe in the acceleration of progress but not in the consequences proposed by Kurzweil
The fiction of the past can become the present of the future; recurrently science "fiction" has become science "fact".
Once again Chomsky has made me evaluate my own preconceived notions about progress and this time in the context of technology. By reminding me computers or robots can be designed to be exceedingly more efficient at repetitive, repeatable uncreative tasks the same way monetary incentive has been found to be successful with human workers with similar work. The same monetary incentive found to be detrimental directed to complete creative problem solving tasks. One then must explore the definitions of intelligence and creativity to decide what would be the definition of sentient creative AI program and would we ever be able to truly declare it having occurred without being sentient computer judges ourselves.
I love how people can't help the urge to attack Noam Chomsky just because they either don't have the capacity or the patience to actually listen to what he is saying. He never claimed to be an expert in artificial intelligence, he just put forth an important argument - based on his area of expertise (linguistics and cognition) - that involves the "hard problem" that we face in trying to understand the nature of thought, much less designing a program to replicate it or surpass it's limitations. I completely agree that ignoring this problem and solely focusing on trivial technological accomplishments (like Watson, Google Translate) today and accepting them as signs of the inevitable fate of the prevalence of ai, and not realizing these to be unable to go beyond the "easy problems" (like computation and organization) due to a missing fundamental piece in our understanding of intelligence, earns ai a rightful spot in the science fiction section in any serious scientific organization.
+Adrian Aguirre The fact Google Translate is considered a success for AI says it all, doesn't it?
Nice interview. Good job Nikola. Thanks.
I was convinced that AI was inevitable. Chomsky makes a very good case that it cannot (not will not) be created.
Maybe later I'll write a rebuttal to some of Noam Chomsky's positions, but for now, I'll just state my disagreement. He clearly hasn't studied AI in depth and is unscientifically biased. Moreover, he's stuck in the past, a radical past I'll grant but the past for sure.
Are you saying that you DO understand how an ant of a certain species in a very hot desert (I forget which) navigates using polarised light. I have never seen such a paper. AI appears to stuck in the past. Even Turing's Test, simple as it seems, has never even been approached. Professor Chomsky mentions this. How do you explain that? We have not even solved the P v NP problem let alone how thinking works. I'd like to hear your "scientific" rebuttal of P. Chomsky.
***** I bring it up because in a paraphrased way that is what prof Chomsky did. When you talk about AI you are presumably talking about human intelligence. I propose that you will never do that because you can't even develop a "theory of mind" for the Ant. You "claim" that it is a trivial problem. It is far from trivial. That's why the only papers in the literature describe the behaviour itself where none decides how the behaviour occurs, mapping what goes on in Ant neurons and synapses, through its central nervous system and so on. Prof Chomsky worked with one single aspect of human intelligence, our ability to speak to each other. That is a very non trivial problem and a glance at his work will show you that. I already have done some reading of the literature in AI and that's the very reason I say I tend to agree with Prof Chomsky's position. I don't simply spout the first thing that comes into my head. Perhaps you won't need to address the P v NP problem but I would suggest that you'll have many much more difficult things to do than that. I have read that current computer scientists DO consider the P v NP problem to be both important and also possibly insoluble with the tools we have. Perhaps we need another leap like the development of the Calculus was a great leap in thinking. I have noticed over and over how you AI fans tend to trivialise what appear to me to be extremely complex ideas, just as you did with my ant or Prof Chomsky's Bee. Finally, I note you're no nearer to "solving" the Turing Test than when Turing posed it. That is a most non trivial test and it was part of Turing's genius to understand why. I have a fairly good grasp of Turing machines, having read "The Emperor's New Mind" by Roger Penrose. It's a brilliant book, replete with the underlying mathematics. I will continue to read the literature in your field but I am not particularly enthused because I really do think you "accentuate the positive and eliminate the negative" in a most unscientific way. Just as you did to Prof Chomsky and far less importantly, to me.
***** We now have very much better understanding of one another's respective positions and clearly we are in broad agreement now. A nice situation to achieve :)
***** Might sound moot at face value, but I would say Chomsky, and even Minsky, would prefer better data and more ways to analyse it than "more data and better ways to analyse it". It's true that there are theories that require higher spatio-temporal resolution in order to be tested, and it's true that the consequent data could fuel further or alternative theories, but such data could potentially reveal the paltry nature of the current theories that forecast and describe the value of that data. Chomsky obviously doesn't think it's likely that "dynamics" would be able to reverse-define the concept of intelligence.
Yes. I was being vaguely tendentious there in my differentiating 'better' and 'more' re data and analysis in that, as I saw Minsky imply elsewhere, maximising efforts in analysis pays forward with an improved economy of data acquisition and processing.
Obviously the human genome project is a pie in the face of that persuasion, but I still find the principle worthy in an age where where data is proliferating and theories are converging.
Obviously high-resolution scanning will pay some dividend, big or small, for investigators of 'cognition'. I just, almost out of cheek, sought to suggest that there are often shitty answers to El Dorado questions (theories). But I have no ambition to even pretend to be specific here.
Great Thoughts!
Always Inspiring!
I put him on my list for future interview requests! ;-)
still true 2022
As Michio Kaku likes to point out, we can't even build an AI as smart as a cockroach. So far we have failed in all attempts to build a machine that can actually learn from the world. The only things machines can do are things we program them to do, and there's no evidence machines can ever evolve past that. All this fear of a singularity and malevolent AIs is jumping the gun, don't you think? It does make for some good science fiction though.
GozerTheGozerian 2 years later and we have build an AI with a small sense of consciousness. It recognises its own voice among others and you can buy the robot for a few hundred dollars
And what's this magic machine called, if I may ask? What you're talking about is smoke and mirrors, just a machine built to mimic conversation, nothing else.
@Darius Wang Chess-playing computers have no bearing on this discussion. Chess is a closed system with rules and a finite number of outcomes. A machine "playing" chess is just doing calculations. That is entirely different from the problem of achieving consciousness in AGI.
Hello Hello That’s not consciousness, that’s the illusion of consciousness. Most of the commentators on here, can’t see the forest from the trees.
Great Interview with a living legend and renaissance man. Well done. I am amazed that he is willing to do this type of interview on the state of the modern world. His opinions are very valuable and we should heed his warnings, and listen to his advice. We need more folks like Dr. Chomsky on earth. Thanks for the video Nikola.
“There are Doctors, and then there are ‘Doctors’ that no machine can replace.”
Can Machines Think?
"I believe to be too meaningless to deserve discussion.”
“Nevertheless, I believe that at the end of the century, the use of words and general educated opinion will have altered so much that one will be able to speak of machines thinking without expecting to be contradicted.”
-Alan Turing (1950) The Imitation Game
Noam is a freakin genius and I love him! However I would humbly say, I think he is a man of his time and his cultural perspective is starting to fall behind. Its not that we need to fix what we have...we need something totally new...Am I crazy? What do you think?
I think Noam is frequently relevant to serious issues that need to be dealt with. Climate change, national debt, income inequality, resource depletion, he gives presentations of the serious issues of our time. Hopefully this interview will wake people up to the fact that there are real issues that deserve concentration. Not fiction about a technical singularity.
Someone else here said it best: Chomsky? I thought he was dead....no, wait...it was just his ideas.
JaxHad I think if you wish to disprove him, you need to come up with a coherent argument, otherwise, yes, you come off as crazy.
Noam doesn't even know to where headphones when doing an interview because of the echo. Brilliant man but technology isn't his strong point. This interview pretty much went as expected.
What an amazing man ! I had only heard of his name, but knew nothing about him. Thank you for the upload.
This is a great interview. Thanks! How did you manage to score an interview with him??
When I hear skepticism about mapping the human brain it reminds me of the genome project.
Can you get Chris Langan on the show?
I'd love to. Do you know how I can get in touch with him?!
Singularity Weblog Yes, I think I can help arrange that. Chris keeps his Email private, so just give me a few weeks of convincing, and I'll get you in touch with him.
Cheers and love your channel
Sounds fantastic and looking forward to hearing back from you. Best way to contact me when you are ready with your update is to do so directly via my contact page: www.singularityweblog.com/services/
Singularity Weblog Awesome, got it. You will be hearing from me in a few weeks then.
Thanks
Singularity Weblog Did Chris Email you yet?
He's only concern is the connection issues, with broadband and what not. He live in the middle of no where. So you guys should iron it out, and we can expect a great show.
cheers
I'm not familiar with that experiment - have you got the name of the article, or a link perhaps?
Great interview. Thank you very much.
His argument against artificial intelligence (at 20:15) seems to be:
- When we talk about intelligent machines, we're talking about intelligent _programs._
- A program is a _theory_. (???)
- The important question about a theory is whether it gives insight and understanding (to whom?) and "these" (which?) don't.
- The real question is how long it will take us to understand intelligence, and we're eons away from that.
This does not sound valid to me.
- I don't see how the distinction between an intelligent program in a device and an intelligent device is important in this context.
- I know what a program is (in software engineering), I know what a theory is (in the physical sciences) and the statement that a program is a theory is nonsense-- unless he's using a special definition of one or both terms, in which case what the hell is he talking about?
- The important thing about a machine (in this context) is not whether it gives us insight, but whether it outsmarts us and perhaps eradicates us.
- The assertion (or implication) that we cannot build an intelligent machine without first having a really solid understanding of intelligence is unproven. We have invented other things that worked quite well before we understood the relevant underlying principles (e.g. animal breeding, water pumps, boomerangs, vaccines, metallurgy, optics, wikipedia), why should intelligence be so different?
+VidkunQL The machine is not creating novel theories (programs) of it's own. It is merely executing novel theories put into it. ie the computer does not CREATE any new programs on it's own outside of the FRAMEWORK of the original program put into it.
It's like how we cannot truly create anything living from non living stuff. Life is essentially molecules which have evolved and learned to do a sort of highly orchestrated dance with each other to achieve certain things, things that even at the most basic level of life are unbelievably complex.
+Jason Williams I have read your comment and considered it carefully; your statements are non-disprovable (because crucial terms like "novel" and "truly create" are undefined) and therefore have no bearing on whether artificial intelligence is possible within the next few decades -- or anything else, for that matter.
I don't mean this as a personal attack, and I think I could convince you that I'm right, provided your livelihood does not depend on believing I'm wrong.
+VidkunQL LMAO Do you even comprehension bruh?
In the simplistic terms "can AI think outside the box ?" or is AI merely a software program designed to spew forth the most logical answer from a database of existing knowledge. Can AI have an imagination to theorize beyond its fundamental scope of knowledge?
To call the concept of the singularity and artificial intelligence 'science fiction' is to basically declare that the human mind operates on magic and therefore cannot be duplicated through applied efforts. Maybe he believes in magic instead of reality. It would seem so based on most of his political ideas.
Thank you. People get so closed-minded and absolute about what is or isn't possible without even realizing it sometimes...
My point is clear: If you declare something to be fiction, you are not suggesting or hinting, hoping or predicting. You are declaring something as if it is factual. The human mind can be figured out as time goes on just like anything else can. For me to say that science will never explain something is the equivalent of saying that it is separate from science - that it is magic. You can't throw around that kind of claim without getting called out on it. Inasmuch as there is no proof that humans will figure out everything, there is no proof that we won't. You can't just say it will never happen and classify future possibilities as fiction. The impossible of yesterday is possible today, through understanding the world better.
He stated that it's scifi, when he doesn't know. I will take it as his opinion. We have to just let the hot air blow sometimes.
FichDichInDemArsch Although I'm not going to take the time to go back and listen to his comments, the assertion that I remember is that human intelligence will forever be impossible to duplicate. That asserts one of the following:
That he knows how far our understanding will come before we stop understanding new things (this would also assert that we will one day stop understanding new things)
That the human brain operates on magic and can never be duplicated with matter and physics.
Intelligence is not subjective or based in any arbitrary opinion. Intelligence is the actual, testable, measurable, observable, duplicatable ability to solve and prevent problems and obstacles. Telling me that intelligence is just someone's perspective makes you wrong.
Thank you so much for posting this.
Excellent interview. Thank you Socrates. Chomsky is absolutely brilliant.
Happy you enjoyed it ;-)
Many OLD people today resist the idea of this kind of thing, even the robots they resist. Their thinking is so linear, so ingrained that they just can't foresee the exponential progression that we have coming in the next few decades.
Clearly someone has been listening to Ray Kurzweil. :)
I recommend you read or listen to his books if you haven't already.
I enjoy listening to Noam, but I really enjoy Ray K.
+sonof hendrix the only things holding us back are the laws of physics!
Glenn Farnan If we live in a simulation (extremely likely) then the laws of the universe could be hacked and changed by AI so we can instantly warp across the universe and create new universes
AI of this era will be and are a load of rules but they are not consciousness or able to creatively reason. That said most medical practitioners could be replaced with a diagnostic algorithm ordering test and proposing treatments.
Still won't you want a human in the loop of your medical treatment? That way you won't be killed by a hard coded error in the algorithms yielding the proposed treatment. Or would you make that treatment required?
The singularity isn't even science fiction, it's science fantasy. You cannot boil down complex choice to binary. As the speaker said, you will not escape the gravity of predetermined bit locations. People like to bring up Star trek tech, while much of it has a plausible grasp on fact, intelligence is infinitely more complex then a warp drive or quantum entanglement, or discovery of exotic particles. Just because processing power of computers increases by hyperthreading and other such methods...
... You cannot boil down complex choice to binary. ... Would it be possible to share more thoughts about this? The issue of the complexity of intelligence reminds me of this saying: It is worse to have a good idea in the hands of the wrong people (stupid, low empathy or simply "very limited"), than to have a bad idea in the hands of the right people (intelligent, ethical or "broadminded"). Singularity, just like "international cooperation" are "nice ideas" and also very "catchy" buzzwords. They can become very different things in the hands of "different" people...
Binary is is basis of all of the highest forms of computing. You cannot change the human code in the machine to overide the machine code when the machine code has the final say. You can put diamonds on a pig, but it's still a pig. You can run enormous constructs of a machine to make it look human, but the final logic groove is slotted in a 1 or 0. Which makes all of its human mimicking moot. Because humans don't have a binary factor to depend on.They have a .05 in their base construct that stalls the execution of idea, but doesn't dismiss the data nor accept it. Computers will always be 100% logic driven, it will never understand the point or purpose of creativity, it will never ask why am I here. These are Human thoughts and concepts developed to work in human minds. Also, the final design to the ultimate machine would find weaker humans a threat to its technological evolution (not sentient) and would find a way to take resource consuming humans that did not have IQ above a certain criteria or provide necessary functions would inherently need to be eliminated from the logic construct of progress.
Fleet Admiral Maybe this is why I could never get any more excited about the subject matter in my required computer course, during my freshman year of undergrad work 30 years ago, than just enough to get a 'C' (or maybe it was a 'B' - I can't even remember). As a music major, I was more intrigued by my guitar and how it gave to me according to how I gave to it. The guitar was made from mostly (though not all) organic material.
Brent, I understand where you are coming from. But just as music works with regards to giving what you get, computers can and do give. It's just not their fault, it is the human that is limited it the concept of sentient life. The only unfortunate part is the assumption by so called bright minds, are nothing more then privileged dreamers.
Computers in themselves, much like musical instruments are tools, and the complexity of the programming is like tuning of an instrument. Yes it would feel more organic as you call it. One of my favorite examples is the song (Me the machine, by Imogen Heap) in which she built a full body computer instrument (see her wired convention talk for info on truly next generation music design).
Also music in itself pays off quickly, and translates easier, but, I think gear like Imogen's sensor gear will make all known instruments obsolete, because of the ability to function as an entire concert ensemble.
That said a quad core is to a single core as metal guitar is to an acoustical guitar.
It's true that bivalent logic has its limits. Have you heard of ternary / trivalent logic? It allows for three truth values: True, False, and Indeterminate. There is no reason to believe that an advanced computer intelligence would limit itself to any one number system upon which to base it's decision making processes. There is already a 9-value logic system designed by the IEEE. With each number system comes new, geometrically increased possible logic gates. Please study geometric growth features in order to fully understand what I'm trying to convey here. Once the logic chaining has evolved to be sufficiently elaborate, it will be identical to "consciousness", human or beyond human. Inevitably this consciousness will grow, far, far beyond what our biological brains are capable of ....
The man is tenured MIT Professor I take his words for it.
appeal to authority
Very notable interview, Nikola. The very fact that Noam Chomsky would agree to discuss a Singulitarian subject matter is testament to its emerging importance. He is calling it "science fiction" in a challenge to the movement's advocates. He is not downing it, he understands the dialectic thesis antithesis synthesis, in the dialectic of human knowledge. He is proposing and articulating the antithesis of a Singularity in order to assist in the proper and humane ramp up to whats gonna happen.