[alicebot-aiethics] Re: alicebot-aiethics digest, Vol 1 #90 - 1 msg
Ryan Ruff
alicebot-aiethics@list.alicebot.org
Mon, 12 Nov 2001 14:29:14 -0800
--------------000800050007060107030300
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
>
>
>>A scarier prospect would be the ability of a bot to
>>hide information from its creator. If the bot is
>>capable of winning the Turing game, it is capable of
>>decieving humans. A bot who can decieve humans and
>>hide the information it learns would become
>>unpredictable. Is this what conscienceness really is?
>>
>
>
>I don't think I've ever read a definition of "consciousness" that
>wasn't, at it's core, equivalent to the definition of "soul". In other
>words, it's a metaphysical/spiritual exercise that cannot be argued
>using logic. Whether or not an AI has a "mind" or "consciousness" or
>"originality" or whatever are questions we must admit we will never be
>able to answer, and they distract from the really important question we
>will need to ask ourselves in the future:
> Should an artificial intelligence be granted legal rights?
>
Personally, I don't subscribe to defining "consciousness" through the
existance of a "soul".
I followed the earlier discussion on the bot as a zombie (september I
believe). The key to understanding zombies is that their only actions
are direct responses to stimuli. If consciousness is the difference
between a zombie and a human then can't we define conciousness as having
the following qualities:
1) The ability to act without being given external stimuli.
2) Unpredictability.
I've tried other definitions, but none to my complete sasifaction. Even
the definition I offer is vague. Most definitions would define
conciousness through a) a soul, b) imagination, or c) emotion.
Defining consciousness by a) requires a spiritual force. Assuming that
even if "God granted humans souls," wouldn't the same God be capable of
granting a machine a soul? In that case, humans would still be no
different from zombies because a zombie could just as easily have a soul.
If we want to use b), we'd have to ask what imagination is. In its
simpliest form, it's the ability to predict the outcom of a chain of
events. We can already program computers to do this.
c) is the most intriguing. Although it seems reasonable, our emotional
responses as humans are an effect of chemical changes that take place
when exposed to stimuli (your body produces adrenaline when you are
scared, etc.). So what's to say that a programed emotional response to
stimuli is any different? The machine is programmed by the creator,
just as we are programmed by our DNA.
>
>We must realize that unless we "give" a bot some sort of legal rights
>(such as Life, Liberty, and Pursuit of Happiness) that prevent us from
>forcibly opening the bot's brain to examination, they will never truly
>be unpredictable because one can, in theory, always open a debug window
>to see the working of it's mind. Or one could simply examine the source
>code.
>
Suppose for a second that an advance form of a MRI could record the
activity of the electrons in the human brain. Further suppose that we
had enough background information to "decode" the thought process that
this electrical activity would produce, thus "debugging" the human brain.
Would this proceduce be legal? Should this procedure be legal?
And if a robot has acheived consciousness (equivolence to humans),
should it be legal to disassembling its code?
>
>Well, okay, I suppose we _could_ build a robot with the following
>characteristics that would by default circumvent all metaphysical *and*
>legal arguments:
> 1) Complete autonomy from human control.
> 2) Invulnerability to probing (through any number of means including
>encryption, overwhelming physical/military might)
>#1 is easy, #2 is hard. Such a robot (even if it failed a Turing Test
>and even if we understood it's brain fully) could in theory insist on us
>giving it legal rights. Regardless of whether or not it has a soul, such
>a robot could force us to treat it with more respect and dignity than
>most humans ever get.
>
I'd argue that neither is actually possible. We, as humans, do not live
in complete autonomy from human control. The belief in such autonomy is
merely an illusion. Every exposure to human activity influences one's
actions. In the case of 2) I refer to my previous hypothetical.
--Ryan
--------------000800050007060107030300
Content-Type: text/html; charset=us-ascii
Content-Transfer-Encoding: 7bit
<html>
<head>
</head>
<body>
<blockquote type="cite" cite="mid:200111121712.MAA11840@wicka.sunlitsurf.com">
<blockquote type="cite">
<pre wrap="">A scarier prospect would be the ability of a bot to<br>hide information from its creator. If the bot is<br>capable of winning the Turing game, it is capable of<br>decieving humans. A bot who can decieve humans and<br>hide the information it learns would become<br>unpredictable. Is this what conscienceness really is?<br></pre>
</blockquote>
<pre wrap=""><!----><br><br>I don't think I've ever read a definition of "consciousness" that<br>wasn't, at it's core, equivalent to the definition of "soul". In other<br>words, it's a metaphysical/spiritual exercise that cannot be argued<br>using logic. Whether or not an AI has a "mind" or "consciousness" or<br>"originality" or whatever are questions we must admit we will never be<br>able to answer, and they distract from the really important question we<br>will need to ask ourselves in the future: <br> Should an artificial intelligence be granted legal rights?<br></pre>
</blockquote>
Personally, I don't subscribe to defining "consciousness" through the existance
of a "soul". <br>
<br>
I followed the earlier discussion on the bot as a zombie (september I believe).
The key to understanding zombies is that their only actions are direct responses
to stimuli. If consciousness is the difference between a zombie and a human
then can't we define conciousness as having the following qualities:<br>
<br>
1) The ability to act without being given external stimuli.<br>
2) Unpredictability.<br>
<br>
I've tried other definitions, but none to my complete sasifaction. Even
the definition I offer is vague. Most definitions would define conciousness
through a) a soul, b) imagination, or c) emotion.<br>
<br>
Defining consciousness by a) requires a spiritual force. Assuming that even
if "God granted humans souls," wouldn't the same God be capable of granting
a machine a soul? In that case, humans would still be no different from
zombies because a zombie could just as easily have a soul.<br>
<br>
If we want to use b), we'd have to ask what imagination is. In its simpliest
form, it's the ability to predict the outcom of a chain of events. We can
already program computers to do this.<br>
<br>
c) is the most intriguing. Although it seems reasonable, our emotional responses
as humans are an effect of chemical changes that take place when exposed
to stimuli (your body produces adrenaline when you are scared, etc.). So
what's to say that a programed emotional response to stimuli is any different?
The machine is programmed by the creator, just as we are programmed by our
DNA.<br>
<br>
<blockquote type="cite" cite="mid:200111121712.MAA11840@wicka.sunlitsurf.com">
<pre wrap=""><br>We must realize that unless we "give" a bot some sort of legal rights<br>(such as Life, Liberty, and Pursuit of Happiness) that prevent us from<br>forcibly opening the bot's brain to examination, they will never truly<br>be unpredictable because one can, in theory, always open a debug window<br>to see the working of it's mind. Or one could simply examine the source<br>code. <br></pre>
</blockquote>
Suppose for a second that an advance form of a MRI could record the activity
of the electrons in the human brain. Further suppose that we had enough
background information to "decode" the thought process that this electrical
activity would produce, thus "debugging" the human brain.<br>
<br>
Would this proceduce be legal? Should this procedure be legal?<br>
And if a robot has acheived consciousness (equivolence to humans), should
it be legal to disassembling its code?<br>
<blockquote type="cite" cite="mid:200111121712.MAA11840@wicka.sunlitsurf.com">
<pre wrap=""><br>Well, okay, I suppose we _could_ build a robot with the following<br>characteristics that would by default circumvent all metaphysical *and*<br>legal arguments:<br> 1) Complete autonomy from human control.<br> 2) Invulnerability to probing (through any number of means including<br>encryption, overwhelming physical/military might)<br>#1 is easy, #2 is hard. Such a robot (even if it failed a Turing Test<br>and even if we understood it's brain fully) could in theory insist on us<br>giving it legal rights. Regardless of whether or not it has a soul, such<br>a robot could force us to treat it with more respect and dignity than<br>most humans ever get.<br></pre>
</blockquote>
I'd argue that neither is actually possible. We, as humans, do not live
in complete autonomy from human control. The belief in such autonomy is
merely an illusion. Every exposure to human activity influences one's actions.
In the case of 2) I refer to my previous hypothetical.<br>
<br>
--Ryan<br>
</body>
</html>
--------------000800050007060107030300--