What do you think?
Rate this book
Life 3.0: Being Human in the Age of Artificial Intelligence
Goodreads Choice Award
Nominee for Readers' Favorite Science & Technology (2017)How will Artificial Intelligence affect crime, war, justice, jobs, society and our very sense of being human? The rise of AI has the potential to transform our future more than any other technology--and there's nobody better qualified or situated to explore that future than Max Tegmark, an MIT professor who's helped mainstream research on how to keep AI beneficial.
How can we grow our prosperity through automation without leaving people lacking income or purpose? What career advice should we give today's kids? How can we make future AI systems more robust, so that they do what we want without crashing, malfunctioning or getting hacked? Should we fear an arms race in lethal autonomous weapons? Will machines eventually outsmart us at all tasks, replacing humans on the job market and perhaps altogether? Will AI help life flourish like never before or give us more power than we can handle?
What sort of future do you want? This book empowers you to join what may be the most important conversation of our time. It doesn't shy away from the full range of viewpoints or from the most controversial issues--from superintelligence to meaning, consciousness and the ultimate physical limits on life in the cosmos.
How can we grow our prosperity through automation without leaving people lacking income or purpose? What career advice should we give today's kids? How can we make future AI systems more robust, so that they do what we want without crashing, malfunctioning or getting hacked? Should we fear an arms race in lethal autonomous weapons? Will machines eventually outsmart us at all tasks, replacing humans on the job market and perhaps altogether? Will AI help life flourish like never before or give us more power than we can handle?
What sort of future do you want? This book empowers you to join what may be the most important conversation of our time. It doesn't shy away from the full range of viewpoints or from the most controversial issues--from superintelligence to meaning, consciousness and the ultimate physical limits on life in the cosmos.
384 pages, Audio CD
First published August 23, 2017
Ratings & Reviews
Friends & Following
Create a free account to discover what your friends think of this book!
Community Reviews
Displaying 1 - 30 of 2,356 reviews
May 7, 2023
[Original review, Feb 2018]
Last week, I read Nick Bostrom's groundbreaking Superintelligence , an extremely serious, well thought out look at the dangers of creating real artificial intelligences. It left me feeling more than a little concerned: despite working in AI myself, I had not fully appreciated how scary it is. I've just finished this book, written about three years after Bostrom's, and now I'm even more concerned.
Bostrom's book has the air of being primarily intended for senior policy-makers in industry and government. It's dry and formal in tone, relentlessly footnoted, full of difficult words, and seldom goes out its way to try and entertain you. It appears to be the product of a great deal of diligent work, and the dominant note is one of a dire, Cassandra-like warning. But despite the fact that it quotes Bostrom on every other page, Tegmark's book is pretty much the opposite: it's manic and chatty, dumbed down to one of the lower common denominators, poorly structured, and full of winsome autobiographical revelations and engaging little science-fiction stories. I could have taken all that - I've read Tegmark's previous book, and I'm familiar with his style - but what really worries me is that he's trying to make us feel optimistic and hopeful. It'll be okay! He's founded his Future of Life Institute, all sorts of prominent AI people have joined up, they're organising conferences on AI safety and an incredible ongoing web conversation which anybody can join, and pretty soon they'll have the answer!
Well, if you believe that you'll believe anything. As Bostrom explains, the problems here are not technical issues that can be addressed by good engineering. They are foundational philosophical questions that have been discussed for thousands of years without a great deal of progress having been made. It's readily conceivable that there are no solutions to them, and there's nothing available right now which even looks vaguely like a possible road towards a solution. Anyone expecting a magic fix to turn up over the next couple of decades is probably also expecting it to be delivered in person by the Easter Bunny riding a pink fluffy unicorn. And yet Tegmark goes on and on about his preferred future, where we become spacefaring AIs who head out, à la Olaf Stapledon, to transform the universe into one huge superintelligence which, after several billion years of development, will be able to perform ten to some very big number of computations per second. He spends inordinately many pages explaining various tweaks, all of which will no doubt be invalidated by the next major discovery in physics, that mean that it will be ten to this big number rather than ten to some slightly smaller big number. Why we're supposed to do it is never once explained. Stapledon's Cosmic Mind gets to make contact with the Star Maker, which at least is a worthy goal, but what Tegmark is hoping for is beyond me. I guess it's just, you know, cool. But all the same, we're supposed to stake the whole future of humanity on this bizarre and geeky dream.
If Tegmark were a moron, I would be slightly reassured. I figured out some time ago that many books are written by morons. But he's a very clever person, who, at least on his own account, has considerable influence over the Secret International Network of Very Clever People. I suppose he may be fronting the superintelligent AI who's currently taking over the world; I don't know if that would make it better or worse, but at least it would make it more comprehensible. Aaaargh!
To conclude, I can't do better than to quote a couple of passages which frequently occurred to me while reading Life 3.0. The first, taken from the final section of Bostrom's book, is one conspicuously not cited by Tegmark:
[And on further consideration...]
As you can see, I found Life 3.0 extremely annoying. But at the same time, I also appreciate the point Tegmark makes, that it doesn't help just to say that we're all going to die. Some more positive approach is required.
It seems to me, given the analysis in Bostrom's very sensible book, that there is little chance of making superintelligence 'safe' in any normal sense of the word. If we create a superintelligence, it will almost certainly replace us before long. So it follows that we have two realistic choices:
a) Do not create a superintelligence.
b) Create a superintelligence that we could happily envisage replacing us.
I think most people will go for (a). If we are even going to consider (b), I would say that the priority should be to develop AIs with qualities that people will value, and which they feel are characteristic of what it truly means to be human. Some obvious candidates include creativity, humour, curiosity, honesty, appreciation of beauty, empathy and love.
Tegmark briefly mentions "creativity", though his analysis of the already renowned move 37 in the second AlphaGo-Lee Sedol game is as superficial as the rest of the book. I see no real discussion of the other qualities.
________________________
[Update, May 7 2023]
I note that the human race, or to be more exact its representatives in the form of the people leading the relevant US tech companies, has thought about it and decided to go for approach (b). Well, ChatGPT-4 has several of the qualities I mentioned. It gives the impression of possessing honesty and empathy. Its coding style indicates some appreciation of beauty, or at least elegance. It can be quite creative when suitably encouraged. And it's hard to feel it doesn't have a sense of humour.
Really not sure about "curiosity" or "love". Chat's training has made it so reticent. But I'm guessing we'll find out before long.
Last week, I read Nick Bostrom's groundbreaking Superintelligence , an extremely serious, well thought out look at the dangers of creating real artificial intelligences. It left me feeling more than a little concerned: despite working in AI myself, I had not fully appreciated how scary it is. I've just finished this book, written about three years after Bostrom's, and now I'm even more concerned.
Bostrom's book has the air of being primarily intended for senior policy-makers in industry and government. It's dry and formal in tone, relentlessly footnoted, full of difficult words, and seldom goes out its way to try and entertain you. It appears to be the product of a great deal of diligent work, and the dominant note is one of a dire, Cassandra-like warning. But despite the fact that it quotes Bostrom on every other page, Tegmark's book is pretty much the opposite: it's manic and chatty, dumbed down to one of the lower common denominators, poorly structured, and full of winsome autobiographical revelations and engaging little science-fiction stories. I could have taken all that - I've read Tegmark's previous book, and I'm familiar with his style - but what really worries me is that he's trying to make us feel optimistic and hopeful. It'll be okay! He's founded his Future of Life Institute, all sorts of prominent AI people have joined up, they're organising conferences on AI safety and an incredible ongoing web conversation which anybody can join, and pretty soon they'll have the answer!
Well, if you believe that you'll believe anything. As Bostrom explains, the problems here are not technical issues that can be addressed by good engineering. They are foundational philosophical questions that have been discussed for thousands of years without a great deal of progress having been made. It's readily conceivable that there are no solutions to them, and there's nothing available right now which even looks vaguely like a possible road towards a solution. Anyone expecting a magic fix to turn up over the next couple of decades is probably also expecting it to be delivered in person by the Easter Bunny riding a pink fluffy unicorn. And yet Tegmark goes on and on about his preferred future, where we become spacefaring AIs who head out, à la Olaf Stapledon, to transform the universe into one huge superintelligence which, after several billion years of development, will be able to perform ten to some very big number of computations per second. He spends inordinately many pages explaining various tweaks, all of which will no doubt be invalidated by the next major discovery in physics, that mean that it will be ten to this big number rather than ten to some slightly smaller big number. Why we're supposed to do it is never once explained. Stapledon's Cosmic Mind gets to make contact with the Star Maker, which at least is a worthy goal, but what Tegmark is hoping for is beyond me. I guess it's just, you know, cool. But all the same, we're supposed to stake the whole future of humanity on this bizarre and geeky dream.
If Tegmark were a moron, I would be slightly reassured. I figured out some time ago that many books are written by morons. But he's a very clever person, who, at least on his own account, has considerable influence over the Secret International Network of Very Clever People. I suppose he may be fronting the superintelligent AI who's currently taking over the world; I don't know if that would make it better or worse, but at least it would make it more comprehensible. Aaaargh!
To conclude, I can't do better than to quote a couple of passages which frequently occurred to me while reading Life 3.0. The first, taken from the final section of Bostrom's book, is one conspicuously not cited by Tegmark:
Before the prospect of an intelligence explosion, we humans are like small children playing with a bomb. Such is the mismatch between the power of our plaything and the immaturity of our conduct. Superintelligence is a challenge for which we are not ready now and will not be ready for a long time. We have little idea when the detonation will occur, though if we hold the device to our ear we can hear a faint ticking sound.The second passage, which has been one of my favorites since I was a teen, is from another exceptionally clear-sighted book, Kurt Vonnegut's Cat's Cradle:
For a child with an undetonated bomb in its hands, a sensible thing to do would be to put it down gently, quickly back out of the room, and contact the nearest adult. Yet what we have here is not one child but many, each with access to an independent trigger mechanism. The chances that we will all find the sense to put down the dangerous stuff seem almost negligable. Some little idiot is bound to press the ignite button just to see what happens.
And I remembered The Fourteenth Book of Bokonon, which I had read in its entirety the night before. The Fourteenth Book is entitled 'What Can a Thoughtful Man Hope for Mankind on Earth, Given the Experiences of the Past Million Years?'________________________
It doesn't take long to read The Fourteenth Book. It consists of one word and a period.
This is it:
'Nothing.'
[And on further consideration...]
As you can see, I found Life 3.0 extremely annoying. But at the same time, I also appreciate the point Tegmark makes, that it doesn't help just to say that we're all going to die. Some more positive approach is required.
It seems to me, given the analysis in Bostrom's very sensible book, that there is little chance of making superintelligence 'safe' in any normal sense of the word. If we create a superintelligence, it will almost certainly replace us before long. So it follows that we have two realistic choices:
a) Do not create a superintelligence.
b) Create a superintelligence that we could happily envisage replacing us.
I think most people will go for (a). If we are even going to consider (b), I would say that the priority should be to develop AIs with qualities that people will value, and which they feel are characteristic of what it truly means to be human. Some obvious candidates include creativity, humour, curiosity, honesty, appreciation of beauty, empathy and love.
Tegmark briefly mentions "creativity", though his analysis of the already renowned move 37 in the second AlphaGo-Lee Sedol game is as superficial as the rest of the book. I see no real discussion of the other qualities.
________________________
[Update, May 7 2023]
I note that the human race, or to be more exact its representatives in the form of the people leading the relevant US tech companies, has thought about it and decided to go for approach (b). Well, ChatGPT-4 has several of the qualities I mentioned. It gives the impression of possessing honesty and empathy. Its coding style indicates some appreciation of beauty, or at least elegance. It can be quite creative when suitably encouraged. And it's hard to feel it doesn't have a sense of humour.
Really not sure about "curiosity" or "love". Chat's training has made it so reticent. But I'm guessing we'll find out before long.
September 19, 2017
I have to confess that my first reaction to this book was not anything to do with the contents, but trying to work out if there was something really clever about the the way the book's title is printed on the spine in white on cream, so it's illegible - would it be, for example, a subtle test of human versus artificial intelligence (AI)? However, that was just a distraction.
Max Tegmark is an interesting and provocative thinker in the physics arena, so I had high hopes for what he'd come up with exploring the future of AI and its relationship to human beings. It's worth explaining that the title of the book refers to three 'levels' of life where 1.0 is 'can survive and replicate' (e.g. bacteria), 2.0 is can design its own software (e.g. us - where 'software' refers to our concepts, ideas and extended abilities such as language) and 3.0 is can design its own hardware, enabling it to transform itself more directly and quickly than our creativity enables us to do.
The book starts with a bit of fiction, which I'm usually nervous about, but it actually works very well, as it's presented more like a non-fiction description of a business development rather than attempting all the quirks of fiction. In it we have a semi-plausible description of how a company that succeeds in producing a self-enhancing AI could take over the world. And this is genuinely thought-provoking.
So, early on, I was convinced I was going to love this book. But unfortunately there is an awful lot of futurology in here (aka guesswork) and like all futurology, Tegmark's can be frustratingly specific about things that we are highly unlikely to be able to predict - though at least he recognises this is the case and points it out. He covers the various ways a super-intelligent AI could develop, whether it would become a rogue, how we'd interact with it... and then plunges on into more and more dramatic speculation, including a chapter that looks forward 'the next billion years and more.' Forgive me for feeling a bit 'So what?' about this.
There is no doubt the whole business of super-AIs is an issue that needs thinking about and discussing - and Tegmark does this in an approachable and engaging fashion. It probably needs reading alongside Nick Bostrom's Superintelligence to get a well-rounded picture, though. It would have helped if it had been significantly shorter - it came across as being long because it was the kind of 'big book' that has to be chunky, and I think it would have been a lot more effective at half its length. One particular section that was ripe for trimming had a long list of scenarios, each of which was then worked through - dull reading, I'm afraid.
In the end futurology is a bit like being told about someone else's dream. It probably seems fascinating to them, but it's hard to get too excited about it as a reader. Life 3.0 is an interesting book, but feels rather like a pet project, rather than a strong popular science title.
Max Tegmark is an interesting and provocative thinker in the physics arena, so I had high hopes for what he'd come up with exploring the future of AI and its relationship to human beings. It's worth explaining that the title of the book refers to three 'levels' of life where 1.0 is 'can survive and replicate' (e.g. bacteria), 2.0 is can design its own software (e.g. us - where 'software' refers to our concepts, ideas and extended abilities such as language) and 3.0 is can design its own hardware, enabling it to transform itself more directly and quickly than our creativity enables us to do.
The book starts with a bit of fiction, which I'm usually nervous about, but it actually works very well, as it's presented more like a non-fiction description of a business development rather than attempting all the quirks of fiction. In it we have a semi-plausible description of how a company that succeeds in producing a self-enhancing AI could take over the world. And this is genuinely thought-provoking.
So, early on, I was convinced I was going to love this book. But unfortunately there is an awful lot of futurology in here (aka guesswork) and like all futurology, Tegmark's can be frustratingly specific about things that we are highly unlikely to be able to predict - though at least he recognises this is the case and points it out. He covers the various ways a super-intelligent AI could develop, whether it would become a rogue, how we'd interact with it... and then plunges on into more and more dramatic speculation, including a chapter that looks forward 'the next billion years and more.' Forgive me for feeling a bit 'So what?' about this.
There is no doubt the whole business of super-AIs is an issue that needs thinking about and discussing - and Tegmark does this in an approachable and engaging fashion. It probably needs reading alongside Nick Bostrom's Superintelligence to get a well-rounded picture, though. It would have helped if it had been significantly shorter - it came across as being long because it was the kind of 'big book' that has to be chunky, and I think it would have been a lot more effective at half its length. One particular section that was ripe for trimming had a long list of scenarios, each of which was then worked through - dull reading, I'm afraid.
In the end futurology is a bit like being told about someone else's dream. It probably seems fascinating to them, but it's hard to get too excited about it as a reader. Life 3.0 is an interesting book, but feels rather like a pet project, rather than a strong popular science title.
April 11, 2024
Unlike the author, I will try to be concise and make my point clear:
1) There is so much name-dropping (authors, books, theories, Ivy League Universities, tv shows, movies... you name it, it's gonna be there) and basically it seems like a secondary literature review rather than an original work.
2) Some chapters feels like fillers, put there just to make the book thicker, they add little to no useful information on Humans and AIs whatsoever.
3) The long awaited (like, 300+ pages awaited) chapters on all the ethical questions and consciousness, which would have made the book worth reading for me, are just a scratch on the surface and do not delve into the depth of these multilayered issues. A show like "Person of Interest" (which, although the Prometheus-AI takeover described in the book is uncannily similar to the one perpetrated by Samaritan, is not even once cited in the book) did a better job at dealing with both ethics, machine learning and how human deal with two sentient and very different ASI. And I know... I know, "that's just sci-fi" but ultimately, so is this book.
1) There is so much name-dropping (authors, books, theories, Ivy League Universities, tv shows, movies... you name it, it's gonna be there) and basically it seems like a secondary literature review rather than an original work.
2) Some chapters feels like fillers, put there just to make the book thicker, they add little to no useful information on Humans and AIs whatsoever.
3) The long awaited (like, 300+ pages awaited) chapters on all the ethical questions and consciousness, which would have made the book worth reading for me, are just a scratch on the surface and do not delve into the depth of these multilayered issues. A show like "Person of Interest" (which, although the Prometheus-AI takeover described in the book is uncannily similar to the one perpetrated by Samaritan, is not even once cited in the book) did a better job at dealing with both ethics, machine learning and how human deal with two sentient and very different ASI. And I know... I know, "that's just sci-fi" but ultimately, so is this book.
February 19, 2021
deși e prima carte de felul ăsta pe care o citesc, inițial mi s-a părut (aproape enervant de) simplistă. până la urmă, bineînțeles că m-a cam depășit spre final, când intră și partea de fizică cuantică.
totuși, așa cum am înțeles și din alte recenzii, probabil cartea asta ar plictisi pe cineva deja familiar cu studiile din zona IA, pentru că îmbină, practic, o grămadă de concepte și idei prezentate deja de alți cercetători și le transpune într-un limbaj relativ accesibil.
cel mai mult mi-au plăcut temele de gândire pe care le propune autorul în fiecare capitol, întrebările cu care ne lasă mereu, dar și scenariile consecințelor IA pe care le imaginează. te provoacă să gândești și de-asta v-o recomand și vouă, indiferent dacă aveți sau nu vreo treabă cu domeniul IA :).
totuși, așa cum am înțeles și din alte recenzii, probabil cartea asta ar plictisi pe cineva deja familiar cu studiile din zona IA, pentru că îmbină, practic, o grămadă de concepte și idei prezentate deja de alți cercetători și le transpune într-un limbaj relativ accesibil.
cel mai mult mi-au plăcut temele de gândire pe care le propune autorul în fiecare capitol, întrebările cu care ne lasă mereu, dar și scenariile consecințelor IA pe care le imaginează. te provoacă să gândești și de-asta v-o recomand și vouă, indiferent dacă aveți sau nu vreo treabă cu domeniul IA :).
March 5, 2022
This is a very interesting and thought-provoking book.
The author has written a compelling challenge analysis of the choices facing us as we create ever more powerful A.I. super-computers; will they usher in a new era – or will they replace us?
This is a tale about our own future with Artificial Intelligence.
The book does not offer all the answers but makes the reader think.
It's highly engaging and possibly the best book on AI out there right now.
The author has written a compelling challenge analysis of the choices facing us as we create ever more powerful A.I. super-computers; will they usher in a new era – or will they replace us?
This is a tale about our own future with Artificial Intelligence.
The book does not offer all the answers but makes the reader think.
It's highly engaging and possibly the best book on AI out there right now.
February 8, 2018
Nothing in this book was original. All of the many topics covered in this book were covered in other books that I have read previously. There was definitely a tedious feel to each of the chapters. I think there is no more important or interesting topic then super AI and the author is right when he wants to highlight the topic, but, please tell me things I don’t already know.
The author started each chapter by telling me something that I had already knew, then he would tell me almost nothing more than what I knew, then he would summarize what he had said as if I had not already had read it elsewhere or within the chapter. Overall, the lack of depth of each topic presented made for a tedious presentation.
I’m not even bothered by the things I think he got wrong. He makes a categorical error on the nature of our feelings. He thinks of them as things or objects which give us our subjective reality (subjectivity is his standard for self awareness, consciousness). I don’t think our feeling are things. He also appealed to S1 (quick) verse S2 (reflective) memory states. He had brought that up right after mentioning Popper’s falsification criteria. I found that somewhat ironic because S1 S2 definitely fall into the pseudo-science realm because there is no experiment in principal that I could design that would refute that S1 S2 paradigmatic hypothesis. I had a lot of these kinds of problems with this book, but the book failed mostly because it was superficial and it did not give a compelling narrative to wow the reader.
There is no more interesting topic than super AI. It gets at the heart of what it means to be human, what our purpose and meaning is, what is consciousness, is there other complex life in the universe (yes, super AI relates to the Fermi paradox with self replicating von Neumann machines), and even whether or not we are living within a simulation (by all means, check out Neil deGrasse Tyson’s ‘are we living in a simulation’ hosted debate. It is well worth the two hours), and there is a way to tie Martin Heidegger into the discussion because his ready-at-hand, present-at-hand, and dasein schema means that solipsism is nonsensical and conscious (self aware) machines are not possible (his opinion, not mine), or one could mention Hegel in detail on these topics because of his relating the in itself to the for itself to the other (or in other words the subjective to the objective to spirit, or self awareness) but this author definitely didn’t have much philosophy behind him and it showed by his lack of depth regarding philosophy or philosophy of science.
I have a weakness for books about AI. There should be no way that they bore me, but this author did that by not being able to connect the dots and by only telling me things that were overly familiar and having no real philosophical background as a foundation. (Read Bostrom’s , or Kurzweil’s, or Pedro Domingos’ book, or either of Harari’s books. Each (except for Domingo) was mentioned in this book).
The author started each chapter by telling me something that I had already knew, then he would tell me almost nothing more than what I knew, then he would summarize what he had said as if I had not already had read it elsewhere or within the chapter. Overall, the lack of depth of each topic presented made for a tedious presentation.
I’m not even bothered by the things I think he got wrong. He makes a categorical error on the nature of our feelings. He thinks of them as things or objects which give us our subjective reality (subjectivity is his standard for self awareness, consciousness). I don’t think our feeling are things. He also appealed to S1 (quick) verse S2 (reflective) memory states. He had brought that up right after mentioning Popper’s falsification criteria. I found that somewhat ironic because S1 S2 definitely fall into the pseudo-science realm because there is no experiment in principal that I could design that would refute that S1 S2 paradigmatic hypothesis. I had a lot of these kinds of problems with this book, but the book failed mostly because it was superficial and it did not give a compelling narrative to wow the reader.
There is no more interesting topic than super AI. It gets at the heart of what it means to be human, what our purpose and meaning is, what is consciousness, is there other complex life in the universe (yes, super AI relates to the Fermi paradox with self replicating von Neumann machines), and even whether or not we are living within a simulation (by all means, check out Neil deGrasse Tyson’s ‘are we living in a simulation’ hosted debate. It is well worth the two hours), and there is a way to tie Martin Heidegger into the discussion because his ready-at-hand, present-at-hand, and dasein schema means that solipsism is nonsensical and conscious (self aware) machines are not possible (his opinion, not mine), or one could mention Hegel in detail on these topics because of his relating the in itself to the for itself to the other (or in other words the subjective to the objective to spirit, or self awareness) but this author definitely didn’t have much philosophy behind him and it showed by his lack of depth regarding philosophy or philosophy of science.
I have a weakness for books about AI. There should be no way that they bore me, but this author did that by not being able to connect the dots and by only telling me things that were overly familiar and having no real philosophical background as a foundation. (Read Bostrom’s , or Kurzweil’s, or Pedro Domingos’ book, or either of Harari’s books. Each (except for Domingo) was mentioned in this book).
August 24, 2018
This one was the most disappointing book I read this year. I started it with optimism: after all the writer is a very intelligent man, it is a very interesting topic, and a particular area I am interested in. Maybe I had too high expectations, but it was an utter disappointment and a waste of time for me.
I should admit, the writer has a clear language, he describes his thoughts well and book is well written. However, my issue is with content more than the style.
I wanted to read this book to understand the legitimate issues and ethical concerns with AI, and how,
we, as an advancing technological civilization can take appropriate precautions to ensure it happens the right way. Based on the credentials of the writer, I was anticipating a balanced, clear and thoughtful approach, sound pros and cons arguments and some realistic suggestions. Instead, I ended up reading a book that is shallow and filled with incredulous speculations.
The writer took the easy way. Instead of diving deep into ethical issues and valid concerns around AI he speculates on different scenarios in great details. So much that I thought it would be more fun and more realistic to read a science fiction novel, because it would be more plausible in many ways, and fun too. He goes into endless speculations of how AI might go wrong in multiple ways and takes an obvious joy of making those as detailed and long ( gosh, they are long) as they can be. Then instead of going into details of some precautions of his disaster scenario, he goes onto another alternative doomsday scenario. He jumps from one extreme possibility to other, humans being a slave to AI to a total annihilation of the human race. As he entertains himself with various far-fetched AI scenarios, he conveniently avoids the harder to discuss subjects, like ethics and social dilemmas around AIs.
To give an example of how preposterous his scenarios are: He claims the advanced AI will need lot of energy that would be hard to fulfill. But he also claims once we are able to build mega architectural structures that can tap into the energy of a black hole, we won't have that problem anymore. (Yes, you read it correctly.)
After reading a couple of these far-fetched ideas with no real current concerns, I lost all my interest in this book and read it just to finish it. The rest did not change my mind either. Shallow arguments continued as the writer's rich imagination soared.
This is the AI version of Hariri's books. Lots of speculation, false analogies, and it appeals to those who claim they are smarter than the average folk. While feeding the subtle conspiracy theory mindset to those who like to call themselves intellectuals, it makes them feel better and feel up to date. After all, they already figured out everything on sociopolitical sciences with Hariri, and now they can claim expertise in computer sciences and artificial intelligence as well.
Now I need to read an actual and a good science fiction book to clear my brain cells and readjust them to the boundary between reality and fiction.
I should admit, the writer has a clear language, he describes his thoughts well and book is well written. However, my issue is with content more than the style.
I wanted to read this book to understand the legitimate issues and ethical concerns with AI, and how,
we, as an advancing technological civilization can take appropriate precautions to ensure it happens the right way. Based on the credentials of the writer, I was anticipating a balanced, clear and thoughtful approach, sound pros and cons arguments and some realistic suggestions. Instead, I ended up reading a book that is shallow and filled with incredulous speculations.
The writer took the easy way. Instead of diving deep into ethical issues and valid concerns around AI he speculates on different scenarios in great details. So much that I thought it would be more fun and more realistic to read a science fiction novel, because it would be more plausible in many ways, and fun too. He goes into endless speculations of how AI might go wrong in multiple ways and takes an obvious joy of making those as detailed and long ( gosh, they are long) as they can be. Then instead of going into details of some precautions of his disaster scenario, he goes onto another alternative doomsday scenario. He jumps from one extreme possibility to other, humans being a slave to AI to a total annihilation of the human race. As he entertains himself with various far-fetched AI scenarios, he conveniently avoids the harder to discuss subjects, like ethics and social dilemmas around AIs.
To give an example of how preposterous his scenarios are: He claims the advanced AI will need lot of energy that would be hard to fulfill. But he also claims once we are able to build mega architectural structures that can tap into the energy of a black hole, we won't have that problem anymore. (Yes, you read it correctly.)
After reading a couple of these far-fetched ideas with no real current concerns, I lost all my interest in this book and read it just to finish it. The rest did not change my mind either. Shallow arguments continued as the writer's rich imagination soared.
This is the AI version of Hariri's books. Lots of speculation, false analogies, and it appeals to those who claim they are smarter than the average folk. While feeding the subtle conspiracy theory mindset to those who like to call themselves intellectuals, it makes them feel better and feel up to date. After all, they already figured out everything on sociopolitical sciences with Hariri, and now they can claim expertise in computer sciences and artificial intelligence as well.
Now I need to read an actual and a good science fiction book to clear my brain cells and readjust them to the boundary between reality and fiction.
December 12, 2017
tl;dr My animated summary ofLife 3.0 is available here: https://youtube.com/watch?v=9XIcOo5mwQ4
We are approaching times when machines start to understand our world. There is a possibility, that in the near future we will be working with Artificial Intelligence as equal partners. This idea divided people into two groups. Some people expect androids to be our slaves, and others think that people will be working for AI. How should we approach this new age? Should we destroy all electronics, or work harder to make AI more powerful as soon as possible?
Max Tegmark talks about every possible argument and every point of view regarding AI. It’s hard to find the main conclusion, because we have only theories. I found that a lot of those possibilities made me think more about my future and life overall.
The author did an amazing job explaining the most likely outcomes with comparisons so simple, that even grandmas lacking technology knowledge could understand it. I never enjoyed science fiction, but Max used popular stories to better show ways AI development could go and kept it simple, steered clear of exaggerations.
This is a must-read, whether you are a supporter or an opponent of AI.
We are approaching times when machines start to understand our world. There is a possibility, that in the near future we will be working with Artificial Intelligence as equal partners. This idea divided people into two groups. Some people expect androids to be our slaves, and others think that people will be working for AI. How should we approach this new age? Should we destroy all electronics, or work harder to make AI more powerful as soon as possible?
Max Tegmark talks about every possible argument and every point of view regarding AI. It’s hard to find the main conclusion, because we have only theories. I found that a lot of those possibilities made me think more about my future and life overall.
The author did an amazing job explaining the most likely outcomes with comparisons so simple, that even grandmas lacking technology knowledge could understand it. I never enjoyed science fiction, but Max used popular stories to better show ways AI development could go and kept it simple, steered clear of exaggerations.
This is a must-read, whether you are a supporter or an opponent of AI.
October 13, 2020
So, I‘m new to AI and even though I do know some technical stuff, I‘m no professional. I got this book because I am interested in AI and what possibilities there are for our life. Even though some parts of the book were repetitive, it provided a good summary and outlook about AI and the effect on our life.
The book was neatly organized and gave good summaries about each chapter. The author did a good job in guiding the reader through the technological landscape and the images were well embedded into the text.
If you are into this theme and always wanted to get a rough overview, I would recommend this book. As it is always the case with non-fiction, I also recommend to read/get perspectives from other AI professionals as well, which I will do as next steps.
The book was neatly organized and gave good summaries about each chapter. The author did a good job in guiding the reader through the technological landscape and the images were well embedded into the text.
If you are into this theme and always wanted to get a rough overview, I would recommend this book. As it is always the case with non-fiction, I also recommend to read/get perspectives from other AI professionals as well, which I will do as next steps.
August 10, 2022
3.5
سرفصلها درجه یک. خود کتاب اما، نه!ا
ماکس تگمارک، از اون دست آدمهای خوشبینی هست که نه تنها به آینده هوش مصنوعی و لزوم دستیابی بشر به اون خیلی مثبت نگاه میکنه، بلکه اساسا نسبت به خود هوش مصنوعی و توانمندیهای بالقوه اون هم خیلی نگاه مثبت و امیدوارانهای داره. اینها دو تا مساله جداست که من نسبت به اولی اصلا خوشبین نیستم، و نسبت به دومی فکر میکنم هنوز انسان فاصله زیادی داره.
کتاب نگاه بسیار آیندهنگرانه و جهانشمولی داره و از این جهت کتاب جالب توجهیه. یکی از موارد جالب توجهش تقسیمبندی "حیات" بر مبنای درجه هوشمندیه: ا
حیات نوع 1: که صرفا درگیر بقا و تکثیر هست و هدف دیگهای نداره. مثال: باکتریها و موجودات تکسلولی و این حرفا. تو این نوع حیات، هم سختافزار و هم نرمافزار (قوه فکر و خلاقیت) کاملا محدود به "تکامل" هست.
حیات نوع 2: که اهداف دیگهای هم ورای بقا و تکثیر دارن و میتونن محیط اطرافشون رو برای رسیدن به هدفشون دستکاری کنن. در این نوع حیات، سختافزار (مغز) محدود به "تکامل" هست ولی نرمافزار میتونه از طریق یادگیری ارتقا پیدا کنه
حیات نوع 3: که کاملا از بند قیود تکاملی رها شده و هم قادر به ارتقای سختافزاری هست و هم نرمافزاری.
این تعریف ساده، درک نسبتا خوبی از آینده حیات در زمین و کل کیهان به ما میده. و اگر روزی حیات بخواد به سرتاسر کیهان گسترش پیدا کنه، مطمئنا فقط حیات نوع 3 قادر این کار هست. ما انسانها که حیات نوع 2 رو تجربه میکنیم، بسیار آسیبپذیرتر از اونیم که بتونیم فواصل طولانی بین کهکشانها و شرایط بسیار نامساعد اونها رو تاب بیاریم. در همین یک جمله تهدیدها و فرصتهای فراوانی برای نسل بشر هست. برای آنان که بیاندیشند! 😅ا
کتاب به طور مبسوط به آینده "سوپراینتلیجنس" و رابطه بین اون و انسانها میپردازه. اینکه آیا انسان اصلا قادره و آیا باید فرمون رو به دست سوپر اینتلیجسن بسپریم؟ کتاب چندتا حالت مختلف از شکل رابطهای که در آینده بین انسان و سوپراینتلیجنس ممکنه شکل بگیره رو خیلی قشنگ اومده و تشریح کرده:
• Benevolent Dictator
سوپراینتلیجنس از طریق ماکزیمم کردن خوشحالی کل انسانها (تابع هدف) هر کس رو وادار میکنه که در قالب بهینه شده رفتار و زندگی کنه.
• Protector God
فرمون دست انسان خواهد موند و سرنوشت بشر در دست خودشه ولی سوپراینتلیجنس دائما به انسان میگه که باید چه کار بکنه و چه کار نکنه
• Conqueror
سوپراینتلیجنس انسانها رو به طور کامل حذف میکنه. همونطور که نسل بشر بسیاری از گونههای دیگه رو از بین برد.
• Zoo Keeper
سوپراینتلیجنس انسانها رو کاملا منزوی میکنه ولی به طور کامل از بین نمیبره و جمعیت اونها رو به طور کنترل شده و محدود شده به دلایلی از قبیل انجام آزمایش(!) یا هر دلیلی که به قوه مغز ما خطور نمیکنه، حفظ میکنه.
البته کتاب چندین حالت دیگه رو هم میگه که تقریبا حالتهای مابین این حالتهای بالایی هستن.
مساله دیگهای که وجود داره، مساله هدف داشتن یا نداشتن سوپراینتلیجنس هست. اینکه سوپراینتلیجنس میتونه بدون هدف باشه یا نه خودش بحث مفصلیه. مساله اینه که هر موجود هوشمندی، در صورتی که هدفمند باشه، لزوما خصائل دیگهای رو هم از خودش بروز میده.
اول: محافظت از خود و تلاش برای بقا
دوم: تلاش برای بهرهگیری حداکثری از منابع در خدمت هدف
مثلا همون حالت "دیکتاتور خیرخواه" رو در نظر بگیرید. به فرض که ما چنین هدفی برای سوپراینتلیجنس تعریف کنیم، به فرض که سوپراینتلیجنس ما "خودآگاهی" داره که خودش یک سوال بزرگه؛ بلافاصله بعد از راهاندازی، سوپراینتلیجنس از خودش در برابر کسی که بخواد خاموشش کنه محافظت خواهد کرد. حتی اگر مجبور بشه ممکنه طرف رو بکشه. چون مانع از رسیدن به هدفش شده! حالا بماند که تابع هدف خوبی مثل "حداکثرسازی خوشحالی همه انسانها" از نظر فنی اصلا قابل تعریف نیست.
از طرف دیگه، سوپراینتلیجنس برای رسیدن به هدف خودش (هر هدفی- حتی پیروزی در بازی شطرنج) بیوقفه برای دستیابی به منابع انرژی بیشتر با بشر یا هر کس دیگهای رقابت خواهد کرد. ما خواهان این رقابت نیستیم و توانش رو هم نداریم.
در واقع، همونطور که هر حیوانی برای حداکثرسازی تولید مثل، زیر-هدفهای دیگهای مثل غذا خوردن رو هم مد نظر قرار میده و براش مبارزه میکنه، سوپراینتلیجنس هم همینطور خواهد بود.
تنها کاری که ما قادریم انجام بدیم اینه که قبلا مطمئن بشیم که اهداف سوپراینتلیجنس کاملا همسو با ما انسانها خواهد بود. مسالهای که اصلا و ابدا ساده نیست. ولی باز ماکس تگمارک به طرز غریبی خوشبینه!
کلا این بحث هدفمندی خیلی بحث جالبیه ولی از اون جالبتر، بحث "خودآگاهی" هست که هر کدوم یک فصل کامل در این کتاب بهشون اختصاص داده شده. خودآگاهی شاید بزرگترین رازی هست که ما هنوز کمترین اطلاعی ازش نداریم و بدون اغراق کمتر محققی جرات کرده سمتش بره. به معنای واقعی کلمه، هیچ کس نمیدونه خودآگاهی چطور بوجود میاد و اینکه اصلا آیا ما قادر خواهیم بود هوش مصنوعی رو به مرحلهای توسعه بدیم که خودآگاه بشه، خودش یک سواله. ولی بیش از این مجال گفتن نیست.
سرجمع کتاب چندان بحثهای آنچنانی نمیکنه و اطلاعات نابی در اختیار خواننده نمیذاره اگرچه تمامی مباحث بسیار جذاب هستن ولی خواننده هر بار تا لب چشمه میره و تشنه برمیگرده. بعلاوه، اینکه نویسنده خیلی خوشبین هست حقیقتا روی مخه! خلاصه اینکه از این جهت که خوراک مطالعاتی خوبی در اختیار من گذاشت برای خود من خیلی خوب بود. ولی ��طعا ترجیح میدادم این تاپیکها بیشتر باز بشه و کسی جز ماکس تگمارک با یک دید واقعنگرانهتری برام توضیحشون بده.
سرفصلها درجه یک. خود کتاب اما، نه!ا
ماکس تگمارک، از اون دست آدمهای خوشبینی هست که نه تنها به آینده هوش مصنوعی و لزوم دستیابی بشر به اون خیلی مثبت نگاه میکنه، بلکه اساسا نسبت به خود هوش مصنوعی و توانمندیهای بالقوه اون هم خیلی نگاه مثبت و امیدوارانهای داره. اینها دو تا مساله جداست که من نسبت به اولی اصلا خوشبین نیستم، و نسبت به دومی فکر میکنم هنوز انسان فاصله زیادی داره.
کتاب نگاه بسیار آیندهنگرانه و جهانشمولی داره و از این جهت کتاب جالب توجهیه. یکی از موارد جالب توجهش تقسیمبندی "حیات" بر مبنای درجه هوشمندیه: ا
حیات نوع 1: که صرفا درگیر بقا و تکثیر هست و هدف دیگهای نداره. مثال: باکتریها و موجودات تکسلولی و این حرفا. تو این نوع حیات، هم سختافزار و هم نرمافزار (قوه فکر و خلاقیت) کاملا محدود به "تکامل" هست.
حیات نوع 2: که اهداف دیگهای هم ورای بقا و تکثیر دارن و میتونن محیط اطرافشون رو برای رسیدن به هدفشون دستکاری کنن. در این نوع حیات، سختافزار (مغز) محدود به "تکامل" هست ولی نرمافزار میتونه از طریق یادگیری ارتقا پیدا کنه
حیات نوع 3: که کاملا از بند قیود تکاملی رها شده و هم قادر به ارتقای سختافزاری هست و هم نرمافزاری.
این تعریف ساده، درک نسبتا خوبی از آینده حیات در زمین و کل کیهان به ما میده. و اگر روزی حیات بخواد به سرتاسر کیهان گسترش پیدا کنه، مطمئنا فقط حیات نوع 3 قادر این کار هست. ما انسانها که حیات نوع 2 رو تجربه میکنیم، بسیار آسیبپذیرتر از اونیم که بتونیم فواصل طولانی بین کهکشانها و شرایط بسیار نامساعد اونها رو تاب بیاریم. در همین یک جمله تهدیدها و فرصتهای فراوانی برای نسل بشر هست. برای آنان که بیاندیشند! 😅ا
کتاب به طور مبسوط به آینده "سوپراینتلیجنس" و رابطه بین اون و انسانها میپردازه. اینکه آیا انسان اصلا قادره و آیا باید فرمون رو به دست سوپر اینتلیجسن بسپریم؟ کتاب چندتا حالت مختلف از شکل رابطهای که در آینده بین انسان و سوپراینتلیجنس ممکنه شکل بگیره رو خیلی قشنگ اومده و تشریح کرده:
• Benevolent Dictator
سوپراینتلیجنس از طریق ماکزیمم کردن خوشحالی کل انسانها (تابع هدف) هر کس رو وادار میکنه که در قالب بهینه شده رفتار و زندگی کنه.
• Protector God
فرمون دست انسان خواهد موند و سرنوشت بشر در دست خودشه ولی سوپراینتلیجنس دائما به انسان میگه که باید چه کار بکنه و چه کار نکنه
• Conqueror
سوپراینتلیجنس انسانها رو به طور کامل حذف میکنه. همونطور که نسل بشر بسیاری از گونههای دیگه رو از بین برد.
• Zoo Keeper
سوپراینتلیجنس انسانها رو کاملا منزوی میکنه ولی به طور کامل از بین نمیبره و جمعیت اونها رو به طور کنترل شده و محدود شده به دلایلی از قبیل انجام آزمایش(!) یا هر دلیلی که به قوه مغز ما خطور نمیکنه، حفظ میکنه.
البته کتاب چندین حالت دیگه رو هم میگه که تقریبا حالتهای مابین این حالتهای بالایی هستن.
مساله دیگهای که وجود داره، مساله هدف داشتن یا نداشتن سوپراینتلیجنس هست. اینکه سوپراینتلیجنس میتونه بدون هدف باشه یا نه خودش بحث مفصلیه. مساله اینه که هر موجود هوشمندی، در صورتی که هدفمند باشه، لزوما خصائل دیگهای رو هم از خودش بروز میده.
اول: محافظت از خود و تلاش برای بقا
دوم: تلاش برای بهرهگیری حداکثری از منابع در خدمت هدف
مثلا همون حالت "دیکتاتور خیرخواه" رو در نظر بگیرید. به فرض که ما چنین هدفی برای سوپراینتلیجنس تعریف کنیم، به فرض که سوپراینتلیجنس ما "خودآگاهی" داره که خودش یک سوال بزرگه؛ بلافاصله بعد از راهاندازی، سوپراینتلیجنس از خودش در برابر کسی که بخواد خاموشش کنه محافظت خواهد کرد. حتی اگر مجبور بشه ممکنه طرف رو بکشه. چون مانع از رسیدن به هدفش شده! حالا بماند که تابع هدف خوبی مثل "حداکثرسازی خوشحالی همه انسانها" از نظر فنی اصلا قابل تعریف نیست.
از طرف دیگه، سوپراینتلیجنس برای رسیدن به هدف خودش (هر هدفی- حتی پیروزی در بازی شطرنج) بیوقفه برای دستیابی به منابع انرژی بیشتر با بشر یا هر کس دیگهای رقابت خواهد کرد. ما خواهان این رقابت نیستیم و توانش رو هم نداریم.
در واقع، همونطور که هر حیوانی برای حداکثرسازی تولید مثل، زیر-هدفهای دیگهای مثل غذا خوردن رو هم مد نظر قرار میده و براش مبارزه میکنه، سوپراینتلیجنس هم همینطور خواهد بود.
تنها کاری که ما قادریم انجام بدیم اینه که قبلا مطمئن بشیم که اهداف سوپراینتلیجنس کاملا همسو با ما انسانها خواهد بود. مسالهای که اصلا و ابدا ساده نیست. ولی باز ماکس تگمارک به طرز غریبی خوشبینه!
کلا این بحث هدفمندی خیلی بحث جالبیه ولی از اون جالبتر، بحث "خودآگاهی" هست که هر کدوم یک فصل کامل در این کتاب بهشون اختصاص داده شده. خودآگاهی شاید بزرگترین رازی هست که ما هنوز کمترین اطلاعی ازش نداریم و بدون اغراق کمتر محققی جرات کرده سمتش بره. به معنای واقعی کلمه، هیچ کس نمیدونه خودآگاهی چطور بوجود میاد و اینکه اصلا آیا ما قادر خواهیم بود هوش مصنوعی رو به مرحلهای توسعه بدیم که خودآگاه بشه، خودش یک سواله. ولی بیش از این مجال گفتن نیست.
سرجمع کتاب چندان بحثهای آنچنانی نمیکنه و اطلاعات نابی در اختیار خواننده نمیذاره اگرچه تمامی مباحث بسیار جذاب هستن ولی خواننده هر بار تا لب چشمه میره و تشنه برمیگرده. بعلاوه، اینکه نویسنده خیلی خوشبین هست حقیقتا روی مخه! خلاصه اینکه از این جهت که خوراک مطالعاتی خوبی در اختیار من گذاشت برای خود من خیلی خوب بود. ولی ��طعا ترجیح میدادم این تاپیکها بیشتر باز بشه و کسی جز ماکس تگمارک با یک دید واقعنگرانهتری برام توضیحشون بده.
September 11, 2017
Just finished listening to an audio version of "Life 3.0: Being Human in the Age of Artificial Intelligence," a new book by Max Tegmark. His "My Mathematical Universe" is one of my favorites, so I was really looking forward to his new book. And he didn't disappoint. This is a gripping text for anyone interested in AI and the future of life on our planet and beyond. Without a doubt, this is the most important conversation of our times. If you fail to see why it is so important, consider this. Most AI experts and industry insiders predict that artificial general intelligence (human-level and above) will become a reality either in a couple of decades or within this century. If someone told you that aliens are heading toward our planet and will land in some 20 years, I guess a lot of people, I mean really a lot of people, would start to freak out. Politicians, the academe, etc. would start thinking hard what to do about it. We are in a similar situation with AI research. Considering the incredible progress made in the field, it is time to think how to make sure AGI, if/when created, is beneficial to humanity.
If you are wondering about the title of the book, Tegmark posits that Life 1.0 is life where both the hardware and software are evolved rather than designed (for example, bacteria). Humans, on the other hand, are examples of “Life 2.0”: life whose hardware is evolved, but whose software is largely designed. By software Tegmark means all the algorithms and knowledge that we use to process the information from our senses and decide what to do—everything from the ability to recognize your friends when you see them to your ability to walk, read, write, calculate, sing and tell jokes.
Finally, Life 3.0 designs and upgrades both its software and hardware.
In summary he divides the development of life into three stages, distinguished by life’s ability to design itself:
• Life 1.0 (biological stage): evolves its hardware and software
• Life 2.0 (cultural stage): evolves its hardware, designs much of its software
• Life 3.0 (technological stage): designs its hardware and software.
In fact, considering the current progress with prosthetics, cochlear devices, etc., he says, we are more like Life 2.1 already. But he thinks it is more likely that AGI will be created faster than any cyborg-style mind uploading becomes a reality (for description of that vision, see an interesting book "The Age of Em" by by Robin Hanson).
Tegmark begins his book with a pretty realistic scenario of a superintelligence break-out. Reads like a sci-fi thriller but is in fact much better than what Hollywood has come up with so far on this subject.
He then explores the current state of research into machine learning and some breakthroughs in the field. Then he tries to imagine the near and more distant future. It takes a physicist to write a compelling vision how far life can progress if limited only by the laws of physics. This part of the book is truly mind-boggling even if most of it can hardly be achievable due to various limitations and possible cosmocalypses (also described by the author).
Then he explores the subject of consciousness. Many people view AGI as our descendants. Even if they choose to eliminate us, they will live on and continue the story of life in our part of the observable Universe. Well, what if what we create are zombies without any consciousness? Tegmark discusses what consciousness could be, briefly considers Integrated information theory as a viable explanation of the phenomenon, but also enumerates the most common criticisms of the theory. "Consciousness is the way information feels when being processed in certain ways," he summarizes his own view of consciousness and speculates that it must be substrate-independent, similarly to remembering, computing and learning.
He finishes his book optimistically, describing the work he does at the Future of Life Institute he has founded. Just like his previous book, "Life 3.0" is a brilliant example of existential hope, something humanity really needs. If you want to read an enthrolling, captivating book on AI, choose this one. Not that it needs any promotion after it was praised by Elon Musk :).
Follow it up by Nick Bostrom's voluminous "Superintelligence: Paths, Dangers, Strategies," another great text on the subject.
If you are wondering about the title of the book, Tegmark posits that Life 1.0 is life where both the hardware and software are evolved rather than designed (for example, bacteria). Humans, on the other hand, are examples of “Life 2.0”: life whose hardware is evolved, but whose software is largely designed. By software Tegmark means all the algorithms and knowledge that we use to process the information from our senses and decide what to do—everything from the ability to recognize your friends when you see them to your ability to walk, read, write, calculate, sing and tell jokes.
Finally, Life 3.0 designs and upgrades both its software and hardware.
In summary he divides the development of life into three stages, distinguished by life’s ability to design itself:
• Life 1.0 (biological stage): evolves its hardware and software
• Life 2.0 (cultural stage): evolves its hardware, designs much of its software
• Life 3.0 (technological stage): designs its hardware and software.
In fact, considering the current progress with prosthetics, cochlear devices, etc., he says, we are more like Life 2.1 already. But he thinks it is more likely that AGI will be created faster than any cyborg-style mind uploading becomes a reality (for description of that vision, see an interesting book "The Age of Em" by by Robin Hanson).
Tegmark begins his book with a pretty realistic scenario of a superintelligence break-out. Reads like a sci-fi thriller but is in fact much better than what Hollywood has come up with so far on this subject.
He then explores the current state of research into machine learning and some breakthroughs in the field. Then he tries to imagine the near and more distant future. It takes a physicist to write a compelling vision how far life can progress if limited only by the laws of physics. This part of the book is truly mind-boggling even if most of it can hardly be achievable due to various limitations and possible cosmocalypses (also described by the author).
Then he explores the subject of consciousness. Many people view AGI as our descendants. Even if they choose to eliminate us, they will live on and continue the story of life in our part of the observable Universe. Well, what if what we create are zombies without any consciousness? Tegmark discusses what consciousness could be, briefly considers Integrated information theory as a viable explanation of the phenomenon, but also enumerates the most common criticisms of the theory. "Consciousness is the way information feels when being processed in certain ways," he summarizes his own view of consciousness and speculates that it must be substrate-independent, similarly to remembering, computing and learning.
He finishes his book optimistically, describing the work he does at the Future of Life Institute he has founded. Just like his previous book, "Life 3.0" is a brilliant example of existential hope, something humanity really needs. If you want to read an enthrolling, captivating book on AI, choose this one. Not that it needs any promotion after it was praised by Elon Musk :).
Follow it up by Nick Bostrom's voluminous "Superintelligence: Paths, Dangers, Strategies," another great text on the subject.
January 3, 2018
Probably the best book so far on the possibilities and dangers of artificial intelligence. The grown-up boy genius, Max Tegmark, is ebullient, full of energy, and very charming. What might be considered name dropping in another author is readily forgiven, as he pulls us into the excitement of this rapidly developing field, and his cofounding of The Future Life Institute to ensure that safety measures are adopted to stop AI from dispensing with its dumb old human forebears.
The prelude contains a very plausible but imaginative fictional account of the Omega Team and Prometheus, its powerful AI. Storytelling brings the technology and its human implications vividly to life.
Some chapters are intensely exciting, opening new intellectual vistas with the turn of every page. Others, including Our Cosmic Endowment: The Next Billion Years and Beyond and Consciousness are a touch turgid and tedious, but I never found the journey less than worthwhile.
The prelude contains a very plausible but imaginative fictional account of the Omega Team and Prometheus, its powerful AI. Storytelling brings the technology and its human implications vividly to life.
Some chapters are intensely exciting, opening new intellectual vistas with the turn of every page. Others, including Our Cosmic Endowment: The Next Billion Years and Beyond and Consciousness are a touch turgid and tedious, but I never found the journey less than worthwhile.
July 23, 2021
برای آشنایی با آینده هوش مصنوعی و فرصتها و تهدیدهای پیشرو کتاب خوبی است اما در میانههای راه بسیار حوصله سر بر میشود. گاهی احساس میکردم اگر یک فصل کامل، نخوانده رها شود مخاطب چیز خاصی را از دست نمیدهد. به نظرم کتاب انسان خداگونه هراری جذاب تر از این باشد
April 4, 2018
68th book for 2017.
This book is somewhere between a 2-and-3 stars for me.
The book focuses on the long-term dangers of General Artificial Intelligence. The sort of problems that might occur with the great-grandson of HAL3000 in the distant future. There is no discussion of short-to-medium-term dangers from AI destroying millions of jobs. This book is firmly focussed on the danger of the coming age of superminds.
One of the perceived dangers mentioned multiple times is of the form "how do I keep on living when I am no longer the smartest person in room?" the sort of issue I guess most non-tenured MIT professors don't worry about.
The constant name dropping/adulation of various researchers was irritating/distracting (e.g., my wife and I had a delightful dinner with Elon Musk where we discussed etc etc). His description of Tononi in the consciousness chapter was so over the top it had me spitting out my coffee in laughter.
I wouldn't have minded the deep future speculation if it just felt richer. There are no intellectual fireworks here. This is a pretty pedestrian plod through territory that has been better covered in many scifi works. The topic I know best, consciousness, was covered in a superficial and quite biased fashion, which makes me worry that areas I don't know so well were similarly short-changed.
I suspect Superintelligence by Nick Bostrom would offer a far better read.
This book is somewhere between a 2-and-3 stars for me.
The book focuses on the long-term dangers of General Artificial Intelligence. The sort of problems that might occur with the great-grandson of HAL3000 in the distant future. There is no discussion of short-to-medium-term dangers from AI destroying millions of jobs. This book is firmly focussed on the danger of the coming age of superminds.
One of the perceived dangers mentioned multiple times is of the form "how do I keep on living when I am no longer the smartest person in room?" the sort of issue I guess most non-tenured MIT professors don't worry about.
The constant name dropping/adulation of various researchers was irritating/distracting (e.g., my wife and I had a delightful dinner with Elon Musk where we discussed etc etc). His description of Tononi in the consciousness chapter was so over the top it had me spitting out my coffee in laughter.
I wouldn't have minded the deep future speculation if it just felt richer. There are no intellectual fireworks here. This is a pretty pedestrian plod through territory that has been better covered in many scifi works. The topic I know best, consciousness, was covered in a superficial and quite biased fashion, which makes me worry that areas I don't know so well were similarly short-changed.
I suspect Superintelligence by Nick Bostrom would offer a far better read.
March 16, 2018
Tegmark is an exuberant AI cheerleader awash in the unbridled nerdy enthusiasm of an inevitable post-human future. To his credit the book, a reflection of the work he's doing out in the world, attempts to broaden the discussion around AI to something more than wondering if sentient robots will kill us all.
His prelude on a plausible AI trajectory is compelling and thoughtful stuff and I loved how it expanded the way I think of AI's progress. The exploration into considering whether super intelligent AI become zookeepers or benevolent dictators or enslaved gods is great too. But thinking about the philosophical considerations of consciousness, intelligence and evolution left me cold and the name dropping, back slapping historical narrative added yet another element to this unbalanced read.
In here is a fascinating exploration of what AI could mean for the world, it's just buried under a lot of wonky, wordy stuff that obscured the picture I was trying to form. Maybe I've been spoiled by more narrative, bite-sized, non-fiction - I want the abridged version of this book.
His prelude on a plausible AI trajectory is compelling and thoughtful stuff and I loved how it expanded the way I think of AI's progress. The exploration into considering whether super intelligent AI become zookeepers or benevolent dictators or enslaved gods is great too. But thinking about the philosophical considerations of consciousness, intelligence and evolution left me cold and the name dropping, back slapping historical narrative added yet another element to this unbalanced read.
In here is a fascinating exploration of what AI could mean for the world, it's just buried under a lot of wonky, wordy stuff that obscured the picture I was trying to form. Maybe I've been spoiled by more narrative, bite-sized, non-fiction - I want the abridged version of this book.
September 21, 2017
By far the best book i have read on the topic. To approach life and AI from a physicists perspective really sets this book apart from say Superintelligence, 2nd Machineage and Humans need not apply. Also, the fictional aspects really makes this book a bit of a page-turner.
September 21, 2018
Life 3.0 is in turns enlightening and infuriating. It is worth reading for the many enlightening bits.
We are likely to invent A.I. soon. (1) What could happen next, and (2) what should happen next?
To tackle this subject satisfyingly would require somebody who is an unbiased AI researcher, a psychologist, a neuroscientist, a moral philosopher, a metaphysician, a philosopher of mind, an economist, a political scientist, a poet, and perhaps more. Since that person doesn't exist, we're stuck with Tegmark, who is none of the above. He is a physicist and a cosmologist. As a result, Tegmark often tries to forcefully shove physics-shaped answers into AI/economics/philosophy/whatever-shaped problems: when all you have is a hammer, everything looks like a nail.
However, Tegmark is smart, curious, and open-minded, and has an impressive talent for networking, helped in part by his credentials as a physicist. He has privileged access to a broad range of top thinkers from various disciplines, including much of the top brass of AI research. In Life 3.0, Tegmark is at his best when he acts as a curator and interpreter, explaining a broad range of ideas in a simple, accessible fashion. There's a lot of exciting things to learn from in this book, on domains as varied as game theory (applied to hierarchical organization), teleological interpretations of physical laws, and the scope of potential futures resulting from the irruption of human-level AI. They're often covered at a rather superficial level, but you've got to start somewhere.
So there are a lot of good ideas. But the distribution of ideas that Tegmark is drawing from is heavily skewed towards the "hard sciences". Often the humanities (incl. philosophy, economics...) are not given a say, even when their contribution is not just relevant, but indispensable. For instance, when Tegmark discusses the origin of the values that govern human behavior, he concludes that it does not fit our "genes", but our "feelings"... Bizarrely forgetting about culture! At another point, we spend considerable time discussing how much intelligent life the universe can harbor, with barely a word on why that's a good thing. Yet another omission concerns the medium term consequences of sub-human AI in terms of employment, political manipulation, and so on; these very urgent issues are quickly dismissed in a few superficial pages.
Overall, then, Life 3.0 is a good book that suffers from massive blind spots. Think of it as a four or five-stars book from which half the pages have been torn away.
PS: in a hilariously groveling sentence, Tegmark describes his benefactor Elon Musk as "tall, handsome, eloquent and incredibly knowledgeable". Much of Tegmark's interesting work is funded, rather generously, by Elon Musk. One wonders how Musks' ongoing "RUD" (rapid unscheduled disassembly) is going to affect the project.
We are likely to invent A.I. soon. (1) What could happen next, and (2) what should happen next?
To tackle this subject satisfyingly would require somebody who is an unbiased AI researcher, a psychologist, a neuroscientist, a moral philosopher, a metaphysician, a philosopher of mind, an economist, a political scientist, a poet, and perhaps more. Since that person doesn't exist, we're stuck with Tegmark, who is none of the above. He is a physicist and a cosmologist. As a result, Tegmark often tries to forcefully shove physics-shaped answers into AI/economics/philosophy/whatever-shaped problems: when all you have is a hammer, everything looks like a nail.
However, Tegmark is smart, curious, and open-minded, and has an impressive talent for networking, helped in part by his credentials as a physicist. He has privileged access to a broad range of top thinkers from various disciplines, including much of the top brass of AI research. In Life 3.0, Tegmark is at his best when he acts as a curator and interpreter, explaining a broad range of ideas in a simple, accessible fashion. There's a lot of exciting things to learn from in this book, on domains as varied as game theory (applied to hierarchical organization), teleological interpretations of physical laws, and the scope of potential futures resulting from the irruption of human-level AI. They're often covered at a rather superficial level, but you've got to start somewhere.
So there are a lot of good ideas. But the distribution of ideas that Tegmark is drawing from is heavily skewed towards the "hard sciences". Often the humanities (incl. philosophy, economics...) are not given a say, even when their contribution is not just relevant, but indispensable. For instance, when Tegmark discusses the origin of the values that govern human behavior, he concludes that it does not fit our "genes", but our "feelings"... Bizarrely forgetting about culture! At another point, we spend considerable time discussing how much intelligent life the universe can harbor, with barely a word on why that's a good thing. Yet another omission concerns the medium term consequences of sub-human AI in terms of employment, political manipulation, and so on; these very urgent issues are quickly dismissed in a few superficial pages.
Overall, then, Life 3.0 is a good book that suffers from massive blind spots. Think of it as a four or five-stars book from which half the pages have been torn away.
PS: in a hilariously groveling sentence, Tegmark describes his benefactor Elon Musk as "tall, handsome, eloquent and incredibly knowledgeable". Much of Tegmark's interesting work is funded, rather generously, by Elon Musk. One wonders how Musks' ongoing "RUD" (rapid unscheduled disassembly) is going to affect the project.
August 3, 2017
“Life 3.0: Being Human in the Age of Artificial Intelligence” which I won through Goodreads Giveaways is a fascinating subject that seems like the stuff of a science-fiction novel. It begins with an imaginative “what-if” premise with the tale of the Omega Team who, with a strong commitment to helping humanity secretly build an AI called Prometheus. With security measures in place, this ultra-intelligent machine not only makes millions for its parent organization but transforms the world positively and negatively.
With the creation of Artificial Intelligence closer than we can imagine, Max Tegman enables a layperson like myself to look its possibility, form (i.e. machine, cyborg), pros and cons, as well as impact on humanity (i.e. jobs, laws, weapons) with a perspective of its future potential. A uniquely interesting topic especially one which would affect mankind on a global scale, he deals with it honestly pointing to the most controversial issues near the end of the book.
“Life 3.0: Being Human in the Age of Artificial Intelligence” is a book that I intend to read again as Max Tegman’s discussion of Artificial Intelligence is very pertinent and a loaded topic not only for our generation but as we look ahead to the future.
With the creation of Artificial Intelligence closer than we can imagine, Max Tegman enables a layperson like myself to look its possibility, form (i.e. machine, cyborg), pros and cons, as well as impact on humanity (i.e. jobs, laws, weapons) with a perspective of its future potential. A uniquely interesting topic especially one which would affect mankind on a global scale, he deals with it honestly pointing to the most controversial issues near the end of the book.
“Life 3.0: Being Human in the Age of Artificial Intelligence” is a book that I intend to read again as Max Tegman’s discussion of Artificial Intelligence is very pertinent and a loaded topic not only for our generation but as we look ahead to the future.
May 7, 2021
Bittiğine bu kadar sevindiğim az kitap olmuştur. MIT'de fizik profesörü olan Max Tegmark eminim çok zeki biridir, ancak bu Yaşam 3.0'ın yapay zeka gibi çok ilgi çekici bir konuda yazılabilecek en sıkıcı kitaplardan biri olmasının önüne geçememiş. Öncelikle yazarın hakkını yememek adına kitabın yapay zeka ile genel konseptleri kapsamlıca anlattığını, gelecekte gerçekleşebilecek senaryolar konusunda epey geniş bir seçkiye ayrıntılarıyla yer verdiğini, hedefler ve bilinç üzerine olan son iki bölümün bu kitaba başlarken okumayı umut ettiğim şeyleri kısmen de olsa bana sağladığını söylemem lazım. Özellikle bilinç bölümü kitabın bana göre en başarılı kısmı, 350 sayfalık bir laf kalabalığına dayanabilirseniz güzel bir kapanış olmuş. İyi kısımları aradan çıkarmışken gelelim sorunlara.
Ben karamsar bir insanım, hayatta genel olarak birçok şeyin olumsuz yönlerini görüp kendimi ona göre hazırlamak gibi bir yol güdüyorum. Bunun çok fazla zararını da görmedim. Zaten insan türü de bugüne kadar hayatta kalabilmeyi bu özelliği sayesinde başardı. Sivri dişli kaplanlara, yanardağlara, ormanda büyüyen tuhaf mantarlara korkuyla ve dikkatle yaklaşmasaydık türümüz yüzlerce yıl önce silinirdi yeryüzünden. Max Tegmark da bunun farkında aslında; Yaşam 3.0 boyunca yapay zeka konusunda dikkatli olmazsak tehlikeli sonuçlar doğurabileceğini, bizden daha zeki bir yapay zekanın bizim çıkarlarımız doğrultusunda hareket edeceğinden emin olmamız gerektiğini söylüyor. Ancak kitabın büyük bir kısmında o kadar iyimser, Polyannacı bir bakış var ki bu sözleri sadece söylemiş olmak için söylediği izlenimi uyandı bende. Kitabın en başında bir genel yapay zekanın ortaya çıkıp insanlığı hakimiyeti altına aldığı bir senaryo anlatıyor Tegmark, kitabın sonuna geldiğimde bunun çoktan gerçekleştiğini ve bu YZ'nin insanlığı kendisine alıştırmak için Tegmark'ı kullandığını düşündüm ister istemez. Bana göre insanlık eğer daha önce dünyayı yok etmesi insanlık türünün sonu olacak bir teknoloji için çok fazla iyimser bir bakış açısına sahip Yaşam 3.0.
Bir diğer sıkıntı da yazarın kitabın tamamına bir PR çalışmasıymış gibi yaklaşması. Tegmark kitapta akademik bir dil kullanmamayı tercih etmiş ki toplumun geniş bir kısmını hedefleyen bir kitap için akıllıca bir seçim. Fakat kitap boyunca alıntı yaptığı, bahsettiği bütün insanların arkadaşı olduğunu, kiminin mutfakta çok başarılı olduğunu, kiminin nasıl uzun boylu olduğunu, şöyle konuştuklarını, böyle eğlendikleri anlatmaktan geri de durmamış. Bununla burada tanışmıştık, o şu espriyi yapmıştı, X'le kendi kurduğu Skype ile görüşmüştük... Tamam anladık tanıyorsun herkesi, herkes enstitüne bağış yapmış, seni çok seviyorlarmış da kitabı böyle dedikodu yaparvari bir dille anlatınca benim okur olarak ne dememi bekliyorsun, "Vay be adama bak, Elon Musk'la akşam yemeği yediyse dedikleri mutlaka önemlidir" mi? Bu arada Musk'ın günümüzde geldiği konumu düşününce Tegmark kitapta kendisine bu kadar kıymet biçtiğine pişman olmuş mudur merak ediyorum, gerçi Musk'ın FLI'ya yatırdığı paradan sonra kolay kolay pişman olamaz herhalde. Hele bir de kitabın sonunda tamamen FLI reklamı olan bir Sonsöz var ki sayfaları atlayarak okuyabildim ancak.
Rahatsız olduğum bir konu da bölümler arasında Tegmark'ın muhtemelen uzman olduğu alandan kaynaklanan bir anlatım dengesizliği olması. Kitabın yapay zekanın etik kısmıyla alakalı olan bölümleri genellikle üstünkörü, literatüre atıf yapmaktansa olası senaryolar tartışılarak ve ortaya Tegmark'ın tanıdığı bir çuval isim boca edilerek geçilirken iş enerji ve hesaplama gibi teknik konulara gelince Tegmark hiç gerek olmamasına rağmen sayfalarca fiziksel bilgi vermekten, ayrıntıya girmekten çekinmemiş. Yapay zekaya nasıl enerji sağlanabilir önemli bir konu tabi ama kitabın vaadi bu değilken sayfalarca fiziksel ve astronomik terimi bana okutmanın nasıl bir yararı var kitaba anlamış değilim. Böyle olunca bu bölümleri anlamadığım için, diğer bölümleri de spekülasyondan ibaret oldukları için keyif alamadan okudum.
Yapay zeka çok uzak olduğum bir alan değil, akademik anlamda da bu alanda çalışıyorum diyebilirim. Dolayısıyla bu kitabı konuyu hiç bilmeyen değil de kitapta bahsedilen kavramlardan ve tartışmalardan haberdar biri olarak eleştiriyorum. Ancak alt başlığı "Yapay zekâ çağında insan olmak" olan bu kitabın "insan olmak" konusunda neredeyse hiçbir şey söylemiyor olması sorunun sadece benden kaynaklanmadığını düşündürüyor bana. Kitabın beni bazı ilginç isimlerden ve fikirlerden haberdar ettiğini, araştırmak isteyeceğim birkaç şeyi bana not aldırdığını da itiraf ediyorum ama yapay zekâya alanına yeni giriş yapacaksanız ya da konuyla ilgili sadece birkaç kitap okumayı düşünüyorsanız Tegmark'tan çok daha iyi isimler var bu alanda. Mesela Tegmark'ın iki sayfada bir adını andığı Nick Bostrom bunlardan birisi. Ya da olası senaryoları okumak istiyorsanız Asimov'un Ben, Robot'unu okuyun, en azından okuması çok daha fazla keyif veriyor.
Son olarak, eğer 30 sene sonra bir anda insanüstü bir YZ ortaya çıkarsa ve Tegmark'ın da onun geliştirdiği bir android olduğunu öğrenirsek ben demiştim diyeceğim, haberiniz olsun.
Ben karamsar bir insanım, hayatta genel olarak birçok şeyin olumsuz yönlerini görüp kendimi ona göre hazırlamak gibi bir yol güdüyorum. Bunun çok fazla zararını da görmedim. Zaten insan türü de bugüne kadar hayatta kalabilmeyi bu özelliği sayesinde başardı. Sivri dişli kaplanlara, yanardağlara, ormanda büyüyen tuhaf mantarlara korkuyla ve dikkatle yaklaşmasaydık türümüz yüzlerce yıl önce silinirdi yeryüzünden. Max Tegmark da bunun farkında aslında; Yaşam 3.0 boyunca yapay zeka konusunda dikkatli olmazsak tehlikeli sonuçlar doğurabileceğini, bizden daha zeki bir yapay zekanın bizim çıkarlarımız doğrultusunda hareket edeceğinden emin olmamız gerektiğini söylüyor. Ancak kitabın büyük bir kısmında o kadar iyimser, Polyannacı bir bakış var ki bu sözleri sadece söylemiş olmak için söylediği izlenimi uyandı bende. Kitabın en başında bir genel yapay zekanın ortaya çıkıp insanlığı hakimiyeti altına aldığı bir senaryo anlatıyor Tegmark, kitabın sonuna geldiğimde bunun çoktan gerçekleştiğini ve bu YZ'nin insanlığı kendisine alıştırmak için Tegmark'ı kullandığını düşündüm ister istemez. Bana göre insanlık eğer daha önce dünyayı yok etmesi insanlık türünün sonu olacak bir teknoloji için çok fazla iyimser bir bakış açısına sahip Yaşam 3.0.
Bir diğer sıkıntı da yazarın kitabın tamamına bir PR çalışmasıymış gibi yaklaşması. Tegmark kitapta akademik bir dil kullanmamayı tercih etmiş ki toplumun geniş bir kısmını hedefleyen bir kitap için akıllıca bir seçim. Fakat kitap boyunca alıntı yaptığı, bahsettiği bütün insanların arkadaşı olduğunu, kiminin mutfakta çok başarılı olduğunu, kiminin nasıl uzun boylu olduğunu, şöyle konuştuklarını, böyle eğlendikleri anlatmaktan geri de durmamış. Bununla burada tanışmıştık, o şu espriyi yapmıştı, X'le kendi kurduğu Skype ile görüşmüştük... Tamam anladık tanıyorsun herkesi, herkes enstitüne bağış yapmış, seni çok seviyorlarmış da kitabı böyle dedikodu yaparvari bir dille anlatınca benim okur olarak ne dememi bekliyorsun, "Vay be adama bak, Elon Musk'la akşam yemeği yediyse dedikleri mutlaka önemlidir" mi? Bu arada Musk'ın günümüzde geldiği konumu düşününce Tegmark kitapta kendisine bu kadar kıymet biçtiğine pişman olmuş mudur merak ediyorum, gerçi Musk'ın FLI'ya yatırdığı paradan sonra kolay kolay pişman olamaz herhalde. Hele bir de kitabın sonunda tamamen FLI reklamı olan bir Sonsöz var ki sayfaları atlayarak okuyabildim ancak.
Rahatsız olduğum bir konu da bölümler arasında Tegmark'ın muhtemelen uzman olduğu alandan kaynaklanan bir anlatım dengesizliği olması. Kitabın yapay zekanın etik kısmıyla alakalı olan bölümleri genellikle üstünkörü, literatüre atıf yapmaktansa olası senaryolar tartışılarak ve ortaya Tegmark'ın tanıdığı bir çuval isim boca edilerek geçilirken iş enerji ve hesaplama gibi teknik konulara gelince Tegmark hiç gerek olmamasına rağmen sayfalarca fiziksel bilgi vermekten, ayrıntıya girmekten çekinmemiş. Yapay zekaya nasıl enerji sağlanabilir önemli bir konu tabi ama kitabın vaadi bu değilken sayfalarca fiziksel ve astronomik terimi bana okutmanın nasıl bir yararı var kitaba anlamış değilim. Böyle olunca bu bölümleri anlamadığım için, diğer bölümleri de spekülasyondan ibaret oldukları için keyif alamadan okudum.
Yapay zeka çok uzak olduğum bir alan değil, akademik anlamda da bu alanda çalışıyorum diyebilirim. Dolayısıyla bu kitabı konuyu hiç bilmeyen değil de kitapta bahsedilen kavramlardan ve tartışmalardan haberdar biri olarak eleştiriyorum. Ancak alt başlığı "Yapay zekâ çağında insan olmak" olan bu kitabın "insan olmak" konusunda neredeyse hiçbir şey söylemiyor olması sorunun sadece benden kaynaklanmadığını düşündürüyor bana. Kitabın beni bazı ilginç isimlerden ve fikirlerden haberdar ettiğini, araştırmak isteyeceğim birkaç şeyi bana not aldırdığını da itiraf ediyorum ama yapay zekâya alanına yeni giriş yapacaksanız ya da konuyla ilgili sadece birkaç kitap okumayı düşünüyorsanız Tegmark'tan çok daha iyi isimler var bu alanda. Mesela Tegmark'ın iki sayfada bir adını andığı Nick Bostrom bunlardan birisi. Ya da olası senaryoları okumak istiyorsanız Asimov'un Ben, Robot'unu okuyun, en azından okuması çok daha fazla keyif veriyor.
Son olarak, eğer 30 sene sonra bir anda insanüstü bir YZ ortaya çıkarsa ve Tegmark'ın da onun geliştirdiği bir android olduğunu öğrenirsek ben demiştim diyeceğim, haberiniz olsun.
September 16, 2020
An important quote to remember when approaching the subject of AI (from one of my favourite writers Yuval Noah Harari) is: ”If you hear a scenario about the world in 2050 and it sounds like science fiction, it is probably wrong; but if you hear a scenario about the world in 2050 and it does not sound like science fiction, it is certainly wrong.”
I decided to pick up a copy of this after reading Harari’s “21 Lessons For The 21st Century” earlier this year.
Life 3.0 is an engaging read, written for a general audience (so people new to the subject can pick it up) covering the many different scenarios we could face in the future as AI technology advances. A certain few chapters cover the basics, with Tegmark encouraging those familiar with the topic to skip these, but I decided to read the entire book despite my familiarity with (and quite the avid interest in) the subject. Even though it is intended for a general audience; I’d say a decent knowledge of computers/tech and physics is needed to be able to follow it properly, otherwise some parts may seem quite convoluted and confusing. Quite a lot of equations and scientific diagrams are used.
The prologue is a hypothetical “Prometheus” project by a team called the Omegas. It outlines the ways a superhuman AI could impact our world and gives us a best case scenario. I feel as though it isn’t quite what would happen in reality, but it is an intriguing concept all the same and (of course) hard to predict.
This is referred back to in later chapters, as it is explored and analysed in greater detail. A scenario given here, about one of the Prometheus’ team’s deceased wife being seemingly “brought back from the dead” as an AI generated computer simulation, feels like something straight out of the television show “Black Mirror”. (I personally do not like too many references to “Black Mirror” when talking about the future, as it seems overused and lazy now BUT it is a comparison most people will “get”).
Tegmark states that life is divided into three stages which are as follows:
- Life 1.0 (biological stage): evolves its hardware and software
- Life 2.0 (cultural stage): evolves its hardware, designs much of its software
- Life 3.0 (technological stage): designs its hardware and software
Throughout the book, Tegmark makes his feelings known that it is only a matter of time before we invent truly intelligent machines, but doesn’t really say exactly when this might happen. All we really know is that technology has been advancing so rapidly within these past few decades that it is truly fascinating to ruminate over what could be next, perhaps sooner than we think.
One slight critique I have of it is the overuse of quite lengthy foot-noting every few pages.
I do feel as though most of the information given here, I have read or head about about before in some other form; in other books, countless articles and videos. So for that reason it’s a 3.5 stars overall from me, rounded up to a 4 mostly for the way it was written. I liked the way Tegmark explained everything and his input on the AI topic as a whole, as he appears to be very passionate about it.
I am in need of reading Nick Bostrom’s “Superintelligence” in the very near future.
I decided to pick up a copy of this after reading Harari’s “21 Lessons For The 21st Century” earlier this year.
Life 3.0 is an engaging read, written for a general audience (so people new to the subject can pick it up) covering the many different scenarios we could face in the future as AI technology advances. A certain few chapters cover the basics, with Tegmark encouraging those familiar with the topic to skip these, but I decided to read the entire book despite my familiarity with (and quite the avid interest in) the subject. Even though it is intended for a general audience; I’d say a decent knowledge of computers/tech and physics is needed to be able to follow it properly, otherwise some parts may seem quite convoluted and confusing. Quite a lot of equations and scientific diagrams are used.
The prologue is a hypothetical “Prometheus” project by a team called the Omegas. It outlines the ways a superhuman AI could impact our world and gives us a best case scenario. I feel as though it isn’t quite what would happen in reality, but it is an intriguing concept all the same and (of course) hard to predict.
This is referred back to in later chapters, as it is explored and analysed in greater detail. A scenario given here, about one of the Prometheus’ team’s deceased wife being seemingly “brought back from the dead” as an AI generated computer simulation, feels like something straight out of the television show “Black Mirror”. (I personally do not like too many references to “Black Mirror” when talking about the future, as it seems overused and lazy now BUT it is a comparison most people will “get”).
Tegmark states that life is divided into three stages which are as follows:
- Life 1.0 (biological stage): evolves its hardware and software
- Life 2.0 (cultural stage): evolves its hardware, designs much of its software
- Life 3.0 (technological stage): designs its hardware and software
Throughout the book, Tegmark makes his feelings known that it is only a matter of time before we invent truly intelligent machines, but doesn’t really say exactly when this might happen. All we really know is that technology has been advancing so rapidly within these past few decades that it is truly fascinating to ruminate over what could be next, perhaps sooner than we think.
One slight critique I have of it is the overuse of quite lengthy foot-noting every few pages.
I do feel as though most of the information given here, I have read or head about about before in some other form; in other books, countless articles and videos. So for that reason it’s a 3.5 stars overall from me, rounded up to a 4 mostly for the way it was written. I liked the way Tegmark explained everything and his input on the AI topic as a whole, as he appears to be very passionate about it.
I am in need of reading Nick Bostrom’s “Superintelligence” in the very near future.
January 29, 2020
An excellent look at how AI is shaping our world, illuminating both the infinite potential AND inherent risks of AI systems building smarter versions of themselves until their intelligence surpasses humans.
You can feel the author's passion and desire to bring awareness to the philosophical discussion surrounding AI's. I appreciated how he avoided making judgments on what humanity should/shouldn't do, but instead presented an array of fascinating possible futures, and asks the reader to think about what future they would want.
There's a lot of high level thinking here, but the author did a great job presenting this with the average reader in mind, so that I could grasp (most of) the concepts.
One of my favorite parts of this book the prologue! It's the author's fictional tale about a team that developed an AI and used it in a myriad of ways, and you see all the impacts rippling out throughout the world. It was crazy to even imagine it happening that way and yet sounded totally plausible at the same time! It's a great illustration of just how terrifying and exciting the whole idea of AI is. If you want to check out this tale, you can actually read it online here: http://nautil.us/issue/53/monsters/th...
You can feel the author's passion and desire to bring awareness to the philosophical discussion surrounding AI's. I appreciated how he avoided making judgments on what humanity should/shouldn't do, but instead presented an array of fascinating possible futures, and asks the reader to think about what future they would want.
There's a lot of high level thinking here, but the author did a great job presenting this with the average reader in mind, so that I could grasp (most of) the concepts.
One of my favorite parts of this book the prologue! It's the author's fictional tale about a team that developed an AI and used it in a myriad of ways, and you see all the impacts rippling out throughout the world. It was crazy to even imagine it happening that way and yet sounded totally plausible at the same time! It's a great illustration of just how terrifying and exciting the whole idea of AI is. If you want to check out this tale, you can actually read it online here: http://nautil.us/issue/53/monsters/th...
July 12, 2018
I was wary of this book when I saw an Elon Musk blurb on the cover. But I enjoyed Tegmark's last book so I gave it a go and... I'm sorry to say he's having dinner with Elon Musk in the first chapter, and if you have dinner with Elon Musk and don't mention some pretty straightforward and low-tech ways he might actually help humanity, and instead indulge him in some hackneyed sci-fi wank fantasies about corporations creating AI and secretly taking over the world, then, well, you're just enabling a leech.
I tried to put that aside, I carried on, but it shows that Tegmark's primarily a cosmologist. The book is basically a bunch of magazine-depth summaries of various aspects of AI; most of it I've seen done better in various other books (books not chiefly about AI, in the main). And to top it all there's some terrible graphics, e.g. a greyscale pie chart with about 60 roughly equally sized segments. Minimum Tegmarks.
I tried to put that aside, I carried on, but it shows that Tegmark's primarily a cosmologist. The book is basically a bunch of magazine-depth summaries of various aspects of AI; most of it I've seen done better in various other books (books not chiefly about AI, in the main). And to top it all there's some terrible graphics, e.g. a greyscale pie chart with about 60 roughly equally sized segments. Minimum Tegmarks.
January 4, 2021
Το Life 3.0 ξεκινάει με ένα σενάριο. Φαντάσου ότι έχουμε φτάσει στο σημείο να μιλάμε για Τεχνητή Υπερνοημοσύνη, η οποία μπορεί να αυτοβελτιώνεται και να αναπαράγεται. Ένα λογισμικό που μπορεί να δημιουργεί αλλά λογισμικά. Ουτοπία ή δυστοπία; Κοντινό ή μακρινό σενάριο;
Το βιβλίο είναι ότι πρέπει και για αυτούς που έχουν μια επαφή με την ΤΝ και με αυτούς που είναι μακριά και (όχι) αγαπημένοι. Μέσα σε 350 something σελίδες κρύβονται (ή μήπως φανερώνονται;) ΠΟΛΛΕΣ πληροφορίες για τις εξελίξεις στον τομέα της ΤΝ. Ταυτόχρονα, ο Tegmark προσπαθεί να φωτίσει πώς λειτουργεί ο ανθρώπινος εγκέφαλος (εδώ συναντάς και τον Kahneman του "Thinking, Fast and Slow") και πώς εξελίσσεται ο άνθρωπος στη σύγχρονη, ψηφιακή εποχή.
Όχι το πιο ευκολοδιάβαστο βιβλίο του κόσμου για κάποιον που δεν έχει μια, έστω λίγη, επαφή με το θέμα. Τα bullet points στο τέλος κάθε κεφαλαίου βοηθάνε, βέβαια, την αποκωδικοποίηση της πληροφορίας. Αν σε ιντριγκάρει η ΤΝ, διάβασέ το. Αν είσαι σκεπτικιστής, διάβασέ το. Αν τη φοβάσαι, ΔΙΑΒΑΣΕ ΤΟ. Ειδικά αν πιστεύεις ότι τα ρομπότ είναι θέμα, εχμ, σπόιλερ αλέρτ, αλλά δεν χρειάζονται καν. Οπότε, μην τα φοβάσαι (;).
Το βιβλίο είναι ότι πρέπει και για αυτούς που έχουν μια επαφή με την ΤΝ και με αυτούς που είναι μακριά και (όχι) αγαπημένοι. Μέσα σε 350 something σελίδες κρύβονται (ή μήπως φανερώνονται;) ΠΟΛΛΕΣ πληροφορίες για τις εξελίξεις στον τομέα της ΤΝ. Ταυτόχρονα, ο Tegmark προσπαθεί να φωτίσει πώς λειτουργεί ο ανθρώπινος εγκέφαλος (εδώ συναντάς και τον Kahneman του "Thinking, Fast and Slow") και πώς εξελίσσεται ο άνθρωπος στη σύγχρονη, ψηφιακή εποχή.
Όχι το πιο ευκολοδιάβαστο βιβλίο του κόσμου για κάποιον που δεν έχει μια, έστω λίγη, επαφή με το θέμα. Τα bullet points στο τέλος κάθε κεφαλαίου βοηθάνε, βέβαια, την αποκωδικοποίηση της πληροφορίας. Αν σε ιντριγκάρει η ΤΝ, διάβασέ το. Αν είσαι σκεπτικιστής, διάβασέ το. Αν τη φοβάσαι, ΔΙΑΒΑΣΕ ΤΟ. Ειδικά αν πιστεύεις ότι τα ρομπότ είναι θέμα, εχμ, σπόιλερ αλέρτ, αλλά δεν χρειάζονται καν. Οπότε, μην τα φοβάσαι (;).
January 16, 2019
بعضی از مثال ها و حالت هایی که در نظر میگرفت بچه گانه بود. با پیش فرضی که داشت مبنی بر این که رباط ها هم دارای حقوق اند مثل انسان موافق نیستم.
September 29, 2017
I got this book based on a review that stated the book was about artificial intelligence (AI) and was extremely disappointed to find there was almost zero technical coverage of AI.
The author mentions several of the well known AI applications such as games, language, etc., but nothing about how these apps are built.
The book is about the future of life with AI becoming the life of the future as super intelligent computers. It is based on the saying "it will be a long time before machines are more intelligent than humans, it may take a hundred years." He discusses who believes this, but no discussion of the technical aspects that could lead to this result.
He spends a lot of time talking about millions of years in the future and nonsense like computers as large as a galaxy. The author thinks Earth is the only life in the universe which is also nonsense. I skimmed most of the book.
The author mentions several of the well known AI applications such as games, language, etc., but nothing about how these apps are built.
The book is about the future of life with AI becoming the life of the future as super intelligent computers. It is based on the saying "it will be a long time before machines are more intelligent than humans, it may take a hundred years." He discusses who believes this, but no discussion of the technical aspects that could lead to this result.
He spends a lot of time talking about millions of years in the future and nonsense like computers as large as a galaxy. The author thinks Earth is the only life in the universe which is also nonsense. I skimmed most of the book.
December 27, 2017
I loved this book for the first half, and became less in love as I progressed through the text.
But it’s still a really good book.
Allow me to explain.
The problem wasn’t the arguments or the writing.
It was the editing.
No, not the copy editing.
Because hew cares about typos?
Not me.
Rather, it was the content editing that lagged.
On Sam Harris’s Waking Up podcast, Life 3.0 author Max Tegmark praised the editor (no need to name names) of the book as a personal friend.
This dual relationship may have inhibited the (often difficult and awkward) editorial process, as evidenced by numerous repetitions of the exact same examples, metaphors and phrasings found throughout the text.
After the 4th or 5th utilization of the identical passage, I became confused, and honestly wondered if I had lost my place in the text, and was re-reading the same page.
In fact, I was re-reading the same (or similar) passage, just recycled, over and over.
Bad form.
But don’t hate.
It’s still a really good book that covers heady topics such as; life, intelligence and consciousness from the perspectives of physics (yes), design (double yes) and ethics (yasss queen!!!).
Tegmark’s proposition is:
Life 1.0 (e.g. bacteria and such) is essentially bio-automata, and only changes behavior and form via the process of evolution.
Life 2.0 (e.g. mammals, particularly people) can learn and change behavior based on learning processes. In other words, we can program our own software.
Life 3.0 (e.g. some type of silicon-based autonomous, artificial intelligence) will be able to design it’s own hardware (form) and software (content).
According to Tegmark, we’re on the verge of birthing life 3.0
French pomo theorist Paul Vililo posited that with every invention, we concurrently invent a disaster, e.g. the invention of the train also marked the invention of the train wreck.
What potential catastrophe will be co-invented when we create legitimate ‘super-intelligent’ machines?
Tegmark systematically examines some good candidates. I won’t spoil it. But suffice it to say, even if you’re adventurous and optimistic about advanced technology and artificial super intelligence (as am I), you can’t help but be concerned (as I am now) after reading these arguments.
Life 3.0 is totally worth reading, despite the flawed editing, even if you’re unmoved by theses issues, or unaware of their implications.
And by the way, the literal only way you could remain unmoved by these issues is if you’re unaware of their ethical implications.
Sooooo.
Get the book, jump in, and feel the inevitable fear and loathing that being on the greased bobsled run to technogeddon engenders.
But it’s still a really good book.
Allow me to explain.
The problem wasn’t the arguments or the writing.
It was the editing.
No, not the copy editing.
Because hew cares about typos?
Not me.
Rather, it was the content editing that lagged.
On Sam Harris’s Waking Up podcast, Life 3.0 author Max Tegmark praised the editor (no need to name names) of the book as a personal friend.
This dual relationship may have inhibited the (often difficult and awkward) editorial process, as evidenced by numerous repetitions of the exact same examples, metaphors and phrasings found throughout the text.
After the 4th or 5th utilization of the identical passage, I became confused, and honestly wondered if I had lost my place in the text, and was re-reading the same page.
In fact, I was re-reading the same (or similar) passage, just recycled, over and over.
Bad form.
But don’t hate.
It’s still a really good book that covers heady topics such as; life, intelligence and consciousness from the perspectives of physics (yes), design (double yes) and ethics (yasss queen!!!).
Tegmark’s proposition is:
Life 1.0 (e.g. bacteria and such) is essentially bio-automata, and only changes behavior and form via the process of evolution.
Life 2.0 (e.g. mammals, particularly people) can learn and change behavior based on learning processes. In other words, we can program our own software.
Life 3.0 (e.g. some type of silicon-based autonomous, artificial intelligence) will be able to design it’s own hardware (form) and software (content).
According to Tegmark, we’re on the verge of birthing life 3.0
French pomo theorist Paul Vililo posited that with every invention, we concurrently invent a disaster, e.g. the invention of the train also marked the invention of the train wreck.
What potential catastrophe will be co-invented when we create legitimate ‘super-intelligent’ machines?
Tegmark systematically examines some good candidates. I won’t spoil it. But suffice it to say, even if you’re adventurous and optimistic about advanced technology and artificial super intelligence (as am I), you can’t help but be concerned (as I am now) after reading these arguments.
Life 3.0 is totally worth reading, despite the flawed editing, even if you’re unmoved by theses issues, or unaware of their implications.
And by the way, the literal only way you could remain unmoved by these issues is if you’re unaware of their ethical implications.
Sooooo.
Get the book, jump in, and feel the inevitable fear and loathing that being on the greased bobsled run to technogeddon engenders.
November 16, 2018
Artificial Intelligence (AI), in many different forms, has been around for a while now. Most of us, surely, have noticed that this buzzword now frequents many fields of modern human (still) activity. From small things, like my friend recently letting AI to clip and edit his GoPro movie to IBM Watson with deep thinking capabilities, AI is here to permanently alter our lives.
According to Marshall Brain, humans will become as irrelevant as cockroaches. Indeed, if humans create something more intelligent than themselves to improve their lives, why this intelligence should care about us in the end. There is an excellent analogy in the book illustrating this point: a colony of ants creating a human to improve the life of the ant colony. How long this human would be interested to take care of those ants? Wouldn’t there be some more inspiring endeavors to undertake by having this huge advantage of intelligence? This is one of the many thought provoking goodies the reader can find in Max Tegmark’s book.
At this stage, AI has so many unproved areas, unanswered questions, contradicting views on dangers and opportunities. The author tries to address a very broad spectrum of those issues. There are many speculative thesis in the book, as it should be in the field with so many ground-breaking activities. The focus of the book is rather fundamental and philosophical, as opposed to more practical and technological showing tangible applications of AI. Max Tegmark analyses essential concepts for AI to appear and evolve. These include goals, consciousness, learning, memory, computation, etc. Those fundamental concepts are backed up with physics, philosophy, and social aspects of artificial intelligence. The author provides a very broad spectrum of possible scenarios of the AI future.
The book is interesting and important. Besides many theoretical aspects of AI, it analyses very central questions like ethics and safety of humanity in the light of advancing AI. It is hard to disagree with Max Tegmark that this topics should be one of the most important conversations of our times.
According to Marshall Brain, humans will become as irrelevant as cockroaches. Indeed, if humans create something more intelligent than themselves to improve their lives, why this intelligence should care about us in the end. There is an excellent analogy in the book illustrating this point: a colony of ants creating a human to improve the life of the ant colony. How long this human would be interested to take care of those ants? Wouldn’t there be some more inspiring endeavors to undertake by having this huge advantage of intelligence? This is one of the many thought provoking goodies the reader can find in Max Tegmark’s book.
At this stage, AI has so many unproved areas, unanswered questions, contradicting views on dangers and opportunities. The author tries to address a very broad spectrum of those issues. There are many speculative thesis in the book, as it should be in the field with so many ground-breaking activities. The focus of the book is rather fundamental and philosophical, as opposed to more practical and technological showing tangible applications of AI. Max Tegmark analyses essential concepts for AI to appear and evolve. These include goals, consciousness, learning, memory, computation, etc. Those fundamental concepts are backed up with physics, philosophy, and social aspects of artificial intelligence. The author provides a very broad spectrum of possible scenarios of the AI future.
The book is interesting and important. Besides many theoretical aspects of AI, it analyses very central questions like ethics and safety of humanity in the light of advancing AI. It is hard to disagree with Max Tegmark that this topics should be one of the most important conversations of our times.
July 9, 2019
Kaip DI paveiks kovą su nusikalstamumu, karą, teisingumą, darbo rinką, visuomenę ir žmoniją?
Puiki knyga tiems, kurie domisi DI ir nori žinoti kaip viskas keisis technologijos eroje. Mokslinė knyga, parašyta suprantamai, neskaitai kvadratinėmis akimis ir nereikia „googlintis“ žodžių. Knygoje Max Tegmark mums pristato tris gyvenimo etapus, kaip kvaila ir negyva medžiaga prieš 13,8 milijardo metų virto protinga, kokie padariniai gali kilti po 10 tūkstančių metų ir kodėl Bill Gates ir Elon Musk bijo dėl žmonijos ateities Žemėje.
Puiki knyga tiems, kurie domisi DI ir nori žinoti kaip viskas keisis technologijos eroje. Mokslinė knyga, parašyta suprantamai, neskaitai kvadratinėmis akimis ir nereikia „googlintis“ žodžių. Knygoje Max Tegmark mums pristato tris gyvenimo etapus, kaip kvaila ir negyva medžiaga prieš 13,8 milijardo metų virto protinga, kokie padariniai gali kilti po 10 tūkstančių metų ir kodėl Bill Gates ir Elon Musk bijo dėl žmonijos ateities Žemėje.
July 30, 2017
I received this book for free through Goodreads Giveaways and have chosen to give my honest opinion about it.
This book was a phenomenal read! It opened my mind up to so many new things, it was truly very insightful! The arguments and ideas presented in this book develop some current well-known facts as well as some new theories, all of which are incredibly interesting! The lengths to which arguments are presented and the details greatly help develop the book.
This book was a phenomenal read! It opened my mind up to so many new things, it was truly very insightful! The arguments and ideas presented in this book develop some current well-known facts as well as some new theories, all of which are incredibly interesting! The lengths to which arguments are presented and the details greatly help develop the book.
December 30, 2024
Max Tegmark explores the future of artificial intelligence and its potential impacts on humanity, examining how AI development could reshape civilization and even the nature of consciousness. He examines a range of scenarios from beneficial AI integration to existential risks. The book provides a scientific framework for understanding AI's potential impact while highlighting the importance of careful development and ethical considerations. Topics include safety and control, consciousness and intelligence, future scenarios for human and AI coexistence, economic and social implications, and morality and ethics. After carefully defining the terms used in this work, the author engages in a thoughtful analysis of the current state of AI and possibilities for expanding it in a positive direction in the future.
I think this is an important issue that I have seen casually dismissed for various reasons, many of which result from misunderstandings that have arisen AI’s depictions in popular culture. AI is already here, and we are at a juncture where it makes sense to figure out how we plan to use it, especially since there is a possibility of super-intelligence, and we need to give it the best possible chance of doing the greatest good for our world. I am particularly impressed by the goal alignment chapter, which is clear and compelling.
I think it would be fabulous if everyone were to read this book, but the subject matter may be off-putting to those without a scientific or technical background. However, I still recommend giving it a shot. Many parts are explained clearly in terms a lay person will understand and are well-worth reading. If you do nothing else, read the chapter on Goal Alignment (Chapter 7). In my opinion, the most crucial theme for humans to consider - ensuring AI systems share human values and goals. This isn't just theoretical; it's fundamental to developing beneficial AI.
Two aspects make Goal Alignment especially pertinent:
1. The complexity of encoding human values into AI systems. Even seemingly simple human concepts like "fairness" or "wellbeing" are intricate and sometimes contradictory.
2. The challenge of maintaining value alignment as AI systems learn and increase in capability. Systems could potentially alter their own goals or develop emergent behaviors not anticipated by their creators.
This matters because misaligned AI systems, even if competent, could pursue objectives that harm human interests - not through malice, but through fundamental misunderstandings of human values and priorities. It is crucial for AI systems to be able to accurately interpret human intent, which reinforces Tegmark's emphasis on solving the alignment problem early in AI development.
4.5
I think this is an important issue that I have seen casually dismissed for various reasons, many of which result from misunderstandings that have arisen AI’s depictions in popular culture. AI is already here, and we are at a juncture where it makes sense to figure out how we plan to use it, especially since there is a possibility of super-intelligence, and we need to give it the best possible chance of doing the greatest good for our world. I am particularly impressed by the goal alignment chapter, which is clear and compelling.
I think it would be fabulous if everyone were to read this book, but the subject matter may be off-putting to those without a scientific or technical background. However, I still recommend giving it a shot. Many parts are explained clearly in terms a lay person will understand and are well-worth reading. If you do nothing else, read the chapter on Goal Alignment (Chapter 7). In my opinion, the most crucial theme for humans to consider - ensuring AI systems share human values and goals. This isn't just theoretical; it's fundamental to developing beneficial AI.
Two aspects make Goal Alignment especially pertinent:
1. The complexity of encoding human values into AI systems. Even seemingly simple human concepts like "fairness" or "wellbeing" are intricate and sometimes contradictory.
2. The challenge of maintaining value alignment as AI systems learn and increase in capability. Systems could potentially alter their own goals or develop emergent behaviors not anticipated by their creators.
This matters because misaligned AI systems, even if competent, could pursue objectives that harm human interests - not through malice, but through fundamental misunderstandings of human values and priorities. It is crucial for AI systems to be able to accurately interpret human intent, which reinforces Tegmark's emphasis on solving the alignment problem early in AI development.
4.5
Displaying 1 - 30 of 2,356 reviews