Showing posts with label Walter Isaacson. Show all posts
Showing posts with label Walter Isaacson. Show all posts

Wednesday, February 11, 2015

Planning the Human-Robot War (Maybe?)

(IMG credit: moviepilot.com)
The (Possibly Kind of Maybe) Rise of the Machines
     What with the recent announcement of a robot-staffed hotel opening in Japan, the internet has been awash with stories of Skynet and robot-human battles for supremacy.  Elon Musk recently donated $10 million dollars toward research designed to prevent a hostile AI from evolving -- and with geniuses of his clout joining institutions like the Singularity University, the public is bound to be nervous.  So just what's at stake?

     Even Hollywood has jumped on board the "singularity" wagon, with the recent Johnny Depp movie Transcendence and the upcoming Terminator reboots (featuring a wise-cracking Arnold Schwarzenegger, and yes I am totally pumped!)  Bestselling tomes on the issue have made millions for authors like the inventor Ray Kurzweil (father of voice recognition, creator of the first working music synthesizer, founder of SingularityU, etc.)  Indeed, it was his book, The Singularity is Near, that gave me my first in depth introduction to the issues and debates at hand -- and outlined Kurzweil's GNR theory of evolution.
IMG: IMDB.com.

     In GNR, intelligent life evolves, and then takes control of its own evolution.  Genetic manipulation comes first, with intelligent life (humans, in our case) being able to recreate their own physical bodies and eliminate disease.  Then comes the nanotechnology revolution (Kurzweil highlights the work of Eric Drexler, considered by many to be the father of nanotechnology theory): any substance will be instantly creaeable, even the human body, and godlike power to manufacture anything, anywhere, in any quantity will be ours.  Finally, both revolutions will play into the Robotics revolution, in which powerful robotics are blended with learning, conscious Artificial Intelligence to create powerful, immortal bodies.  At this point, the Singularity occurs -- according to Kurzweil, humanity will reach an undefinable point of evolution where matter, space, and even time itself are utterly manipulable.

     Granted, further research revealed that many of these admittedly exciting ideas are familiar from popular culture.  The Singularity/godhood event are familiar from Arthur C. Clarke and Stanley Kubrick's collaboration, 2001: A Space Oddyssey (read the book first, it's better).  Indeed, the most recent mindbender from the dark and brilliant mind of Christopher Nolan, Interstellar, features similar ideas (no spoilers here, go watch it for yourself!  It's brilliant, stimulating, and annoying, like most of Nolan's work).
Interstellar - IMG: theguardian.com

     Perhaps most intriguingly, Kurzweil-esque theories seem to match up fairly well with what we've pieced together of evolution (from Dawkins et al) -- RNA finds a way to replicate, in theory, causing abiogenesis (as Kurzweil emphasizes, all matter is patterns, and intelligent or self-organizing patterns are more powerful).  It is this theme of growing ability to organize that really stands out about the whole GNR idea -- there's an air of rightness about it, a feeling that this was meant to be.   The brilliant writers and speakers all seem to know what they're talking about, and the ramifications are so startling and exciting that you want to believe and fear not to do so.

     But powerful objections remain.  One of the most powerful comes from Jaron Lanier, a programmer who argues that humans are grossly incapable at programming, which means that programming a learning, self-aware AI is impossible, or will at the very least take a VERY long time.  Others, like biologist PZ Myers, scathingly accuse Kurzweil and others of lazily glossing over the vast biological (esp. neurological) complexities involved in any "reverse-engineering" of the brain and thus consciousness.  We simply don't know enough to even predict the nature of future evolution.
Img: mcb.berkely.edu

     Finally, the most compelling (and comforting) argument that I have encountered synthesizes both viewpoints (an approach I always appreciate).  Yes, computers are advancing rapidly -- at one way of thinking, argues Walter Isaacson in The Innovators, his masterwork on the history of computing.  Isaacson posits that computers and humans are good at different and complimentary tasks, and while computers can grow infinitely in their niche, they will never be able to match the power of humans and computers working in tandem.  He backs this up with hundreds of pages that trace the symbiotic nature of computers and humans, which was the original intention of many of the masterminds of the computer revolution (like JCR Licklider and Vannevar Bush).

     So in the end, while it's always fun to kick back and watch Terminator, and while it's prudent to create safeguards against technology gone amok, and while it is further pleasant to await potential enhancements to human existence -- don't hold your breath.  Make the most of the gifts of the age, be aware of possible danger, but no matter what, LIVE HUMANLY AND HUMANELY.  Even if the meaning of the first word evolves in the next century! :)

Thursday, February 5, 2015

Random Balm for the Digital Soul: The Innovators by Walter Isaacson (and other chaotic detritus)

Well, every once in a while you need a nice nerdly spree.

Mine started perhaps 12 hours ago, when I sat down and resolved that if I was going to spend miserable hours or days waiting to hear back from prospective employers, I might as well go ahead and do my best to be productive in the meanwhile.  And/or enjoy myself.

Whether any of these things actually happened, is (of course) open to interpretation and debate.  In the meantime, I shall use the excellent advice of a very wise blogger and author: PERFECTION DELAYS ACTION.  This has become my mantra and shall help to motivate me in posting yet another obscure and nearly unread blog post.  Who knows - maybe in the year 2100, some poor graduate research assistant will be helping to research or ghostwrite a book on 21st century madmen and will dig up my post on one out of a billion blogs and decide (hope against hope) that I produced prolific words of wisdom, a veritable golden horde of tasty metaphysical tidbits.

Maybe I need to go back to see that shrink.

In any case, I was delighted to finish (among other things) the marvelous The Innovators by Walter Isaacson, an excellent, broad, and deep history of computing, starting with Ada Lovelace and Charles Babbage and going all the way through to the present day.  You can read my review on Amazon.

The main joy for me, besides FINALLY getting a decent overview of the rise of the (for me) intimidating Information Age, and besides being shown the beauty of the intersection of arts and sciences (at last, I understand what Hitchens was trying to say in years of essays), I received a full overview of Artificial Intelligence and the competing views on the future of intelligent life on the planet.

You see, when reading brilliant authors like Ray Kurzweil, you can quickly despair and see a future in which humans become obsolete (don't get me wrong, Arthur C. Clarke painted a beautiful vision of this in 2001: A Space Oddyssey 50 years before Kurzweil began penning his bestselling nonfiction discussions of the putative Singularity).  I like Isaacson's far more nuanced view of a symbiotic relationship between humans and computers, as exemplified with the work IBM has done with Watson (one of the world's premier super computers).  "In other words, the future might belong to people who can best partner and collaborate with computers," he writes (p. 476).  Isaacson describes a researcher who helped the medical community use Watson for researching cancer, and her words are powerful:

"'I watched Watson interact in a collegial way with the doctors," she said. 'It was the clearest testament of how machines can truly be partners with humans rather than try to replace them.  I feel strongly about that.'"

Isaacson follows the narrative with a heart-warming alternative to what Jaron Lanier has (somewhat unfairly) called Cybernetic Totalism (the idea that humans will be replaced/destroyed by intelligent machines):

"Perhaps no matter how fast computers progress, artificial intelligence may never outstrip the intelligence of the human-machine partnership."  (P. 478).

Even if human intelligence (H) remains constant, the nearly infinite growth of computer intelligence (C) will, Isaacson contends, always trump mere infinite computer intelligence; in my handy-dandy dummy's equation (I have been spending quality time with Asimov's Physics, after all): HC>C.  As Bender would say in Futurama, Neat!

Isaacson's lengthy book convinces me that this is a more likely scenario, especially after detailing the many predictions and actual failures of true artificial consciousness to materialize, and detailing the long history of symbiosis or cybernetic thought throughout the history of computing innovations (see the sections on Vannevar Bush and his memex, or JCR Licklider and his philosophies).

(Yes, I know that consciousness has yet to be satisfactorily defined, and that it is possible that we ourselves are not exhibiting independent thought, etc. etc.  I haven't finished Dennett's book, and I'll probably never understand it, but I feel comfortable with the layman's understanding of the term!)

In other news, I have no idea why AdSense is giving me the middle finger (but shall continue valiantly to fill the void with the meaningless ravings of an unstable and sleep-deprived mind), I'm still torn between my respect for Maajid Nawaz and my instinctive antipathy for religion in general, I need to start eating better before I actually have a heart attack, and I shall always remember with great fondness the day a kindhearted Lisa Shearin answered a tweet and one of my favorite authors answered a good natured email in kind.

How can I recover my old sense of humor and lose the unwelcome grimness?

For amazing and incredible advice on breaking out of unemployed misery after a liberal arts degree:

http://www.selloutyoursoul.com/
http://blog.aftercollege.com/

Cheers!