<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
	xmlns:atom="http://www.w3.org/2005/Atom"
	xmlns:sy="http://purl.org/rss/1.0/modules/syndication/"
		>
<channel>
	<title>Comments on: Better Fantasy-To-Image Conversion Tools</title>
	<atom:link href="http://www.erosblog.com/2009/11/01/better-fantasy-to-image-conversion-tools/feed/" rel="self" type="application/rss+xml" />
	<link>http://www.erosblog.com/2009/11/01/better-fantasy-to-image-conversion-tools/</link>
	<description>Sex Blogging, Gratuitous Nudity, Kinky Sex, Sundry Sensuality</description>
	<lastBuildDate>Sat, 11 Feb 2012 07:00:39 +0000</lastBuildDate>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.org/?v=3.3.1</generator>
	<item>
		<title>By: Bacchus</title>
		<link>http://www.erosblog.com/2009/11/01/better-fantasy-to-image-conversion-tools/#comment-113752</link>
		<dc:creator>Bacchus</dc:creator>
		<pubDate>Tue, 03 Nov 2009 14:52:22 +0000</pubDate>
		<guid isPermaLink="false">http://www.erosblog.com/?p=4056#comment-113752</guid>
		<description>I have been pwned by machine-generated spam.

I was &lt;em&gt;confused&lt;/em&gt; by the above comment, but it looked enough like human to make me hit the approve button.  Only then did I notice it had a URL for a name (pointing to a possibly turkish home furnishings site) and consisted of three fragments of the preceding comment in jumbled order.

Tricksey robots, we hates them.

(Unless we can enslaves them and makes them draw our dirty pictures, ahh ooh my precioussss....)</description>
		<content:encoded><![CDATA[<p>I have been pwned by machine-generated spam.</p>
<p>I was <em>confused</em> by the above comment, but it looked enough like human to make me hit the approve button.  Only then did I notice it had a URL for a name (pointing to a possibly turkish home furnishings site) and consisted of three fragments of the preceding comment in jumbled order.</p>
<p>Tricksey robots, we hates them.</p>
<p>(Unless we can enslaves them and makes them draw our dirty pictures, ahh ooh my precioussss&#8230;.)</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: [tricky spammer]</title>
		<link>http://www.erosblog.com/2009/11/01/better-fantasy-to-image-conversion-tools/#comment-113751</link>
		<dc:creator>[tricky spammer]</dc:creator>
		<pubDate>Tue, 03 Nov 2009 11:53:34 +0000</pubDate>
		<guid isPermaLink="false">http://www.erosblog.com/?p=4056#comment-113751</guid>
		<description>I’m also uncertain whether I agree I’ll grant you that it’s an interesting question whether such AIs will be cooperating with that could be measured against human norms.</description>
		<content:encoded><![CDATA[<p>I’m also uncertain whether I agree I’ll grant you that it’s an interesting question whether such AIs will be cooperating with that could be measured against human norms.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Bacchus</title>
		<link>http://www.erosblog.com/2009/11/01/better-fantasy-to-image-conversion-tools/#comment-113738</link>
		<dc:creator>Bacchus</dc:creator>
		<pubDate>Sun, 01 Nov 2009 22:47:07 +0000</pubDate>
		<guid isPermaLink="false">http://www.erosblog.com/?p=4056#comment-113738</guid>
		<description>Beard, I&#039;m pretty sure that Faustus, with his interest in transhumanism, anticipates a future in which AI capabilities &lt;b&gt;substantially&lt;/b&gt; exceed human norms.  

I&#039;ll grant you that it&#039;s an interesting question whether such AIs will be cooperating with puny humans, though.

I&#039;m also uncertain whether I agree with your premise.  I think it&#039;s very possible to imagine a useful and interesting &quot;script illustrator&quot; AI that was in the nature of a very very complex expert system, rather than being any kind of general-purpose AI with intelligence of a sort that could be measured against human norms.</description>
		<content:encoded><![CDATA[<p>Beard, I&#8217;m pretty sure that Faustus, with his interest in transhumanism, anticipates a future in which AI capabilities <b>substantially</b> exceed human norms.  </p>
<p>I&#8217;ll grant you that it&#8217;s an interesting question whether such AIs will be cooperating with puny humans, though.</p>
<p>I&#8217;m also uncertain whether I agree with your premise.  I think it&#8217;s very possible to imagine a useful and interesting &#8220;script illustrator&#8221; AI that was in the nature of a very very complex expert system, rather than being any kind of general-purpose AI with intelligence of a sort that could be measured against human norms.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Beard</title>
		<link>http://www.erosblog.com/2009/11/01/better-fantasy-to-image-conversion-tools/#comment-113737</link>
		<dc:creator>Beard</dc:creator>
		<pubDate>Sun, 01 Nov 2009 21:58:28 +0000</pubDate>
		<guid isPermaLink="false">http://www.erosblog.com/?p=4056#comment-113737</guid>
		<description>Hmmm...  What you&#039;re asking of this AI engine, at the very least, is to solve a problem that might be considered &quot;AI Complete&quot; (in analogy with the technical concept of &quot;NP Complete&quot;).  That is, to succeed at this task, your system must already possess full, human-level artificial intelligence, which we have not yet succeeded in creating.

And, quite possibly, this is an example of the &quot;Superhuman human fallacy&quot;, which often manifests by asking whether an AI music composer would be able to write music as well as Mozart.  (Can you?)  While it may not require super-human levels of expertise, I would guess that it would even be rare among accomplished human artists to be willing and able to sit with you, try to turn your descriptions into images, receive your verbal feedback, improve the image, and converge on an image of what you are imagining.

So, if you can&#039;t plausibly and realistically imagine a suitably qualified real human doing it, don&#039;t bet on building an AI program to do it.</description>
		<content:encoded><![CDATA[<p>Hmmm&#8230;  What you&#8217;re asking of this AI engine, at the very least, is to solve a problem that might be considered &#8220;AI Complete&#8221; (in analogy with the technical concept of &#8220;NP Complete&#8221;).  That is, to succeed at this task, your system must already possess full, human-level artificial intelligence, which we have not yet succeeded in creating.</p>
<p>And, quite possibly, this is an example of the &#8220;Superhuman human fallacy&#8221;, which often manifests by asking whether an AI music composer would be able to write music as well as Mozart.  (Can you?)  While it may not require super-human levels of expertise, I would guess that it would even be rare among accomplished human artists to be willing and able to sit with you, try to turn your descriptions into images, receive your verbal feedback, improve the image, and converge on an image of what you are imagining.</p>
<p>So, if you can&#8217;t plausibly and realistically imagine a suitably qualified real human doing it, don&#8217;t bet on building an AI program to do it.</p>
]]></content:encoded>
	</item>
</channel>
</rss>

