<br><br><div><span class="gmail_quote">On 05/06/07, <b class="gmail_sendername">Eugen Leitl</b> <<a href="mailto:eugen@leitl.org" target="_blank" onclick="return top.js.OpenExtLink(window,event,this)">eugen@leitl.org</a>
> wrote:<br></span><br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
Weapon design is not merely an intellectual problem, and neither do<br>theoretical physicists operate in complete detachment from the empirical<br>folks. I.e. the sandboxed supergenius or braindamaged idiot savant is a<br>
synthetic scenario which is not going to happen, so we can ignore it.</blockquote><div><br>It might not happen with humans, because they suffer from desires, a bad temper, vanity, self-doubt, arrogance, deceitfulness etc. It's not their fault; they were born that way. But why would anyone deliberately design an AI this way, and how would an AI acquire these traits all by itself?
<br></div><br></div><br>-- <br>Stathis Papaioannou