15 November 2012

Early take on Kurzweil's 'How to Create a Mind'

Came across 'Ray Kurzweil's Dubious Theory of Mind' posted today by Gary Marcus at the New Yorker site today.  Marcus, a professor of psychology at NYU, does a pretty thorough job of tearing down Kurzweil's book, and from all I know about Kurzweil I have to agree.  As is pointed out in the posting, Kurzweil's done some amazing things in his life, but at the same time he frequently glosses over or simplifies things that deserve a lot more thought and effort.  Here's an excerpt:
At the beginning of the book, Kurzweil promises to reverse engineer the human brain in hopes of using the brain’s secrets to advance artificial intelligence, but what he’s really done is the opposite: reverse engineer his own companies’ computer systems in order to propose a theory about how the mind works.

Ultimately Kurzweil is humbled by a challenge that has beset many a great thinker extending far beyond his field—Kurzweil doesn’t know neuroscience as well as he knows artificial intelligence, and doesn’t understand psychology as well as either. (And for that matter he doesn’t know contemporary A.I. as well as the A.I. of his heyday, when he was running his companies thirty years ago.)
I'll still take a look at the book (from the library! doesn't sound like a keeper), but I'm not sure Kurzweil is adding much this time around.

I've written about Kurzweil before: see here and here.

No comments: