As part of my summer reading on the topic of Artificial Intelligence, I finished Ray Kurzweil's book, "How to create a mind." Good start to get me thinking about this topic.
Kurzweil covers a lot of ground in this short book. Everything from the latest understanding on how brain works, to psychology to philosophy. And of course computing, computer science and how all of this comes together. I was a bit worried that he might proceed from there into a quasi-techno-religious direction - but thankfully did not.
One of the key concepts that Kurzwiel keeps coming back to in the book is the Turning test.
I suppose what I’m trying to get at is at what point in time on a grade of hybridization is Surya-bot no longer Surya?
Kurzweil covers a lot of ground in this short book. Everything from the latest understanding on how brain works, to psychology to philosophy. And of course computing, computer science and how all of this comes together. I was a bit worried that he might proceed from there into a quasi-techno-religious direction - but thankfully did not.
The word that leaped out for me in the book was confabulation. That
is, we mostly tend to confabulate, rather than straight out lie (the
latter having deceitful intent). Confabulation is also a medical
condition for people with dementia.
I just love this word and have been mulling it...
It jokingly occurs to me that when two wise people meet, this is what happens:
Wise person 1: silence
Wise person 2: silence (with a smile)
I played off off the Turing test
concept that Kurzweil highlighted in the book, coming up
playfully with the "inverse Turing test."
In a Turing Test, a human sits across a computer separated by a wall. The human has to tell if the dialogue is with a computer or a human. The computer passes the Turing Test if the human quizzing it can not tell the difference (picturized beautifully in the movie Ex Machina)
In a Turing Test, a human sits across a computer separated by a wall. The human has to tell if the dialogue is with a computer or a human. The computer passes the Turing Test if the human quizzing it can not tell the difference (picturized beautifully in the movie Ex Machina)
Here is the revision to the Turning test I was playing around with.
What if you know
on the other side of the wall from you is Surya .... and I gradually, step by
step become a hybrid. (ie someone you know and thus specifically can relate to the person you are interacting with).
Here it is, my revised Turing Test
1.
At some point, as we grow old if I had a knee replacement would you
still engage with me as the Surya you know and are comfortable with?
(Yes / No)
2. How about in addition if I put in a ear implant?
(Yes / No)
3. How about if I also got a kidney replaced?
(Yes / No)
4.
How about when I get to an age where my memory is failing. And with advanced
technology available many years from now, I get a memory aid implant?
Would you still interact with me like same old same old?
(Yes / No)
5.
How about if I lose my speech capability but am able to interact with
you through technology (like Stephen Hawkins did through the speech
synthesizer in real life)?
(Yes / No)
6.
Suppose I am further handicapped but have the assistance of an A.I.
which knows all my thought patterns and has access to all my life
experiences, all my photos and my writings.
And suppose it sends online messages as Surya-bot. Also it creates replies and chats on my behalf but only posts them only if I say ‘yes” to each message it would send.
Would you interact with Surya-bot like you interacted with Surya prior to the incident in question#1?
(Yes / No)
I suppose what I’m trying to get at is at what point in time on a grade of hybridization is Surya-bot no longer Surya?
----------
May the bots behave!

6 comments:
Here is a native version: origin unknown.
A: " This is my Grandfather's pocket knife"
B: "Wow! seems to be in good condition!"
A" "Yes! I replaced the blade thrice and the handle twice."
This is a very complex aspect of the human mind Surya - the perception of the human mind of other individuals.
It seems to have entirely different set of parameters for creating personality profiles of immediate family members, relatives, friends, business contacts, chance acquaintances etc.
So your question may have multiple answers.
Love it!!
I read super intelligence. It’s sobering!!!
Thank you Surya (hopefully not Surya-bot).
This is a fascinating topic, and frankly quite scary in some ways.
I guess that as it gradually changes we will become more and more open to over time (as your illustration below).
My answers to your questions are: Yes, across the board.
Hmmm… I suppose it’d be when you ask me why I should be allowed to exist, given the uselessness of such failed carbon-based life as humans.
Or perhaps, being too smart to ask such a question, when you suggest that there’s a wonderful Disneyland like place where I can spend my life having fun without the need to work or trouble myself with such silly existential questions like a search for meaning or purpose….
Post a Comment