Review of 'What to think about machines that think' on 'Storygraph'
2 stars
A few gems among many essays that seem to have little original or useful in them.
I've read several of these Edge essay collections. This is the worst. It may be that this interesting topic is too complex to say anything useful in a page or two which is the normal length in this book. A few are much shorter, a few spill over into a third page. One of the short ones looks like the author was declining the invitation to contribute (roughly, "I don't think that machines think, so I don't have much to say").
Too many of the essays just go over the same ground. Either
"I define thinking narrowly, so based on my definition, these purported examples of machine thought don't qualify" or
"I define thinking broadly, and here are some examples of machines doing it"
Too many just give opinions of what will be possible …
A few gems among many essays that seem to have little original or useful in them.
I've read several of these Edge essay collections. This is the worst. It may be that this interesting topic is too complex to say anything useful in a page or two which is the normal length in this book. A few are much shorter, a few spill over into a third page. One of the short ones looks like the author was declining the invitation to contribute (roughly, "I don't think that machines think, so I don't have much to say").
Too many of the essays just go over the same ground. Either
"I define thinking narrowly, so based on my definition, these purported examples of machine thought don't qualify" or
"I define thinking broadly, and here are some examples of machines doing it"
Too many just give opinions of what will be possible in the future, without supporting evidence or even logic. A few at least introduce the opinion with "I suspect that", so that we know they're just gut feelings that we can ignore.
I read this on Kindle. I went back through my highlights. Only a handful were of the "interesting, think about this some more" type. Most were "here's another example of why this deserves a low rating.
Many of the thinkers, in spite of their excellent reputation in their own field, are not experts in this field and some of them are not aware of recent advances in the field. The experts generally do a good job of explaining where some marketing claims are just hype, so if you are not aware of the state of the art, you may learn something, but you'd do just as well with some Google searches.
