Superintelligence: Philosophy with a Deadline

In the relatively nascent field of artificial intelligence, few books have had the impact that ‘Superintelligence’ by Nick Bostrom has had.  Bostrom is a Swedish philosopher at Oxford University and in a short space of time has become one of the most respected thinkers in the field.  His book is a sweeping journey through the various risks posed to society by the prospect of superintelligence - taking on each risk with the technical focus of a computer scientist, the philosophical deftness of an ethicist and the psychological skepticism of a sociologist.  This combination of viewpoints made the book a fundamental read for anyone interested in the field.  It’s widely accepted as a masterpiece.

So, as someone looking to work in the field, I picked up the book and ploughed through it.  And now I sit here in front of my laptop trying to summarise my thoughts.  I look to my notes next to me and for the life of me can’t find a way to put them together in a way that would make sense to someone who hasn’t read the book.  It’s a weird feeling.  Most books, especially in the non-fiction section, have a few key ideas which can be summarised and delivered in a blog post form.  With this one however, I don’t think I would do it justice by doing that.  In addition, to discuss one or two ideas out of context of the rest of the book would do more harm than good, I think.  The immense breadth and depth of topics covered in the book each deserve their own treatises and that’s the reason why this book has become so foundational for the field.

So while I know this is going to be immensely unsatisfying for you, the reader, my only recommendation is for you to read the book from front to back yourself.

If you want to understand the predicament we sit in right now, with advancing machine intelligence gathering more and more speed, then this is a book you must read.  It’s not an easy book to read but I think it’s a crucially important one.  Artificial intelligence will change this world and it’s vitally important that we develop the technology in a way that promotes our own best interests - to deliver on the tremendous potential upsides and steer away from the existential disasters that await if we are asleep at the wheel.