Scan barcode
muksreads's review against another edition
3.0
This book is hugely important but written in a way which is difficult for the average reader. I find it hard to imagine how to deliver the information in this book in an easily readable format but it was still an issue, one that the author himself acknowledges.
Despite this I find the content of the book fascinating. Bostrom does a great job at underlining how superintelligence may not be as far away as one might imagine and how we could possibly deal with the emergence of one, before and after its creation. The subject of this book does make it rather difficult to read. I do not believe it is meant to be consumed in one sitting, there are complex topics, dystopian ideas and constant new terminology thrown at you.
I do not reget reading this at all, it has helped me understand superintelligence more and I would strongly recommend with a warning: the author does not attempt to frame the discussion in a particularly negative or positive light, he simply writes it as it is to the best of his ability, and that can be frightened.
Despite this I find the content of the book fascinating. Bostrom does a great job at underlining how superintelligence may not be as far away as one might imagine and how we could possibly deal with the emergence of one, before and after its creation. The subject of this book does make it rather difficult to read. I do not believe it is meant to be consumed in one sitting, there are complex topics, dystopian ideas and constant new terminology thrown at you.
I do not reget reading this at all, it has helped me understand superintelligence more and I would strongly recommend with a warning: the author does not attempt to frame the discussion in a particularly negative or positive light, he simply writes it as it is to the best of his ability, and that can be frightened.
josephcooper's review against another edition
challenging
dark
informative
reflective
sad
slow-paced
5.0
jimmyceroneii's review against another edition
3.0
More Philosophy than Tech
I loved the big questions this book asked as I think it revealed some important issues with Superintelligence that I hadn’t considered. That said, this book could have been 50 pages shorter. The author tried to cover all his bases and it was utterly overwhelming - I barely finished the book.
I loved the big questions this book asked as I think it revealed some important issues with Superintelligence that I hadn’t considered. That said, this book could have been 50 pages shorter. The author tried to cover all his bases and it was utterly overwhelming - I barely finished the book.
cynicusrex's review against another edition
4.0
This is no trivial read. I'm convinced I'll give this book five stars after reading it again.
w1nd0w's review against another edition
I honestly don't care what this man has to say about AI and Superintelligence anymore because we have hit a point where his understanding is meaningless.
thefriedone's review against another edition
4.0
Dense but fascinating look at artificial intelligence and the next step, superintelligence. Bostrom has written the to-date definitive textbook on the topic.
We learn about different approaches researchers might take to achieving artificial intelligence, the pros and cons of each approach, and the odds that each will be the first to get there. We also learn about how these early stage AIs will most likely evolve (quicker than you might expect) into a superintelligence (essentially a digital being so much smarter than us that it will easily take over the world, perhaps even within hours of coming into being). Finally, because of the dire consequences involved, Bostrom discusses various ways we might try to prevent a superintelligence from taking over the world and imbue it with values that will make it cooperate with humans by design.
The first few chapters read like a computer science text, but the bulk here is more philosophy than hard science. What values/goals of an AI would most likely lead it to work side-by-side with humanity instead of destroying us? How do we get it to follow these values? How should humanity cooperate with each other in the process leading up to the AI revolution, considering the fact that the first success in this field may quickly determine the fate of Earth and the whole universe?
While dense at times, this is a fascinating read.
If you're interested in the topic but don't want to dive into the deep end just yet, check out this great, long blog series that covers the same topics and cites plenty of Bostrom's work: http://waitbutwhy.com/2015/01/artificial-intelligence-revolution-1.html
We learn about different approaches researchers might take to achieving artificial intelligence, the pros and cons of each approach, and the odds that each will be the first to get there. We also learn about how these early stage AIs will most likely evolve (quicker than you might expect) into a superintelligence (essentially a digital being so much smarter than us that it will easily take over the world, perhaps even within hours of coming into being). Finally, because of the dire consequences involved, Bostrom discusses various ways we might try to prevent a superintelligence from taking over the world and imbue it with values that will make it cooperate with humans by design.
The first few chapters read like a computer science text, but the bulk here is more philosophy than hard science. What values/goals of an AI would most likely lead it to work side-by-side with humanity instead of destroying us? How do we get it to follow these values? How should humanity cooperate with each other in the process leading up to the AI revolution, considering the fact that the first success in this field may quickly determine the fate of Earth and the whole universe?
While dense at times, this is a fascinating read.
If you're interested in the topic but don't want to dive into the deep end just yet, check out this great, long blog series that covers the same topics and cites plenty of Bostrom's work: http://waitbutwhy.com/2015/01/artificial-intelligence-revolution-1.html
itsokayitsofficial's review against another edition
3.0
First things first: this is NOT a "popular science" read a la Malcolmn Gladwell, Neil DeGrasse Tyson, or other scilebrity who's name I can't spell either.
This is structured like a research proposal and reads like a text book. Admittedly an unusually well written and engaging text book, but such is a low bar to begin with. It's dense, often repetitive, and frustratingly exhaustive when it doesnt need to be, while not when it should be.
That being said still a good read and definitely the best that speaks on the topic.
This is structured like a research proposal and reads like a text book. Admittedly an unusually well written and engaging text book, but such is a low bar to begin with. It's dense, often repetitive, and frustratingly exhaustive when it doesnt need to be, while not when it should be.
That being said still a good read and definitely the best that speaks on the topic.