<div class="gmail_quote">On 24 January 2012 00:30, Anders Sandberg <span dir="ltr"><<a href="mailto:anders@aleph.se">anders@aleph.se</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
But much of biological cognition is strongly hardware dependent<br clear="all"></blockquote></div><br>I am quite skeptical as the existence of forms of cognition, or any other information processing for that matter, that would be "hardware dependent" at all.<br>
<br>The real question remains however of what kind of performance hit you are going to suffer by running some of, or all, the relevant routines on "hardware" which can be much, much less optimised to do so.<br><br>
This is why, BTW, in spite of steampunk fantasies to the contrary we do not seriously consider the development of AGIs on punch-card machines.<br><br>-- <br>Stefano Vaj<br>