Thus, the internet and our ability to search it are within reach of the limits of the raw storage and communications capacity of the human brain, and should exceed it by 2015.[Perhaps due to page coding problems, the link to Sejnowski doesn't bring me to the right page. If you want to read his article, click here and scroll a little way down and you'll find Sejnowski]
If I understand Sejnowski correctly, he suggests that the Internet might achieve sentience sometime after 2015. That's in 9 years! : )
Sejnowski's Dangerous Question reminded me of the Arnold Schwarzenegger Terminator movies, where they mention SkyNet becoming sentient and starts taking over the world. Internet... SkyNet... creepy coincidence or deliberate reference?
As a Sci Fi fan, I tend to think about stuff like "Machine Sentience and Implications for Humanity". Whenever such themes are explored in movies, more often than not, the Machines are set up for confrontation against Humans.
What's becoming clear to me is that the really Dangerous Idea isn't about the Internet achieving awareness per se. What's perceived as danger could be the perceived impact of machines being self-conscious.
Think about it -- just being "aware" doesn't necessarily mean becoming violent. Maybe the Post-awareness Internet will be a benevolent presence, kicking the butts of spammers and email scammers.
The really Dangerous Idea might be this: "A Post-Awareness Internet inherently spells doom for humans".
Even without articulating it, most people will subsconsciously assume that any man-made construct that becomes 'alive' is automatically a threat. I don't think it has to do with robots only. Moreover, implicit in the term "Machine Sentience" is that Machines are more advanced and developed than Humans.
I think Humans are hardwired to be wary of Competition, and the instinctive reaction is to eliminate competition, more so when we think the other person is better than us. Survival of the fittest and all that. Perhaps jealousy is also something hardwired into us also. We can't stand to have more than one Sentient Being share the same planet. We're not big-hearted enough for that.
However, that's not to say there isn't hope for humanity. What humans have shown -- in fact it's a survival trait -- is that our behaviour can be changed. And that humans are perhaps the most adept in consciously modifying our behaviour.
So we have maybe 9 years to learn how to live with (the possibility of) another Sentient Being.
Meantime, I guess we can start by being nicer to the next Sentient Being that's already here with us.
Other Human Beings.
Tag: machine intelligence
Hi Ivan,
ReplyDeleteThis is a fascinating topic. For a more in-depth look at it, I would highly recommend Ray Kurzweil's books The Age of Spiritual Machines and The Singularity is near. (First for readability, second for depth.)
If you are a sci-fi fan, there is no end, of course, to the novels, films, etc., that explore this topic. But a very powerful treatment is given by Dan Simmons in his 4 book Hyperion series. The AI "Core" is very powerful indeed.
Thanks for this entry!
Doug
This is something I've been fascinated about since the creation of the Internet. I fear less of the network automatically becoming conscious, but more of the line between life and artificial intelligence crossing over.
ReplyDeleteIn brief, if we can make web services (e.g. recommendation services) to military robots mimick our better judgement through algorithms, what do we end up with if a blind test reveals that we can no longer tell the responses of a living human from the pseudo-living machine?
My point is exacted in the 2nd installment of Ghost in the Shell 2, which is out as anime movie on DVD and a novel.
How do we tell when we've crossed that line?
Thanks Doug, I'll certainly look up the reading recommendations.
ReplyDeleteKevin -- tough question lah. My answer: "We Don't. 'Cos it'll be them who'll tell us".