wedrifid comments on Nonperson Predicates - Less Wrong

28 Post author: Eliezer_Yudkowsky 27 December 2008 01:47AM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Article Navigation

Comments (162)
Tags: naturalism seq_fun fai consciousness

Comments (162)

Sort By: Old

You are viewing a single comment's thread. Show more comments above.

Comment author: ArisKatsaris 17 November 2011 01:59:59AM *  2 points [-]

If there's an ethical issue with killing simulations there's an ethical issue with killing AIs.

Doesn't follow, for several reasons:

  • If the issue is with the termination of subjective experiences, and if we assume that that people-simulations have qualia (let's grant it for the sake of argument) it still doesn't follow that every optimization algorithm of sufficient calculational power also has qualia.
  • If the ethical issue is with violation of individuals' rights, there's nothing to prevent us from constructing only AIs that are only too happy to consent to be deleted; or indeed which strongly desire to be deleted eventually -- but most people-simulation would presumably not want to die, since most people don't want to die.
Comment author: wedrifid 17 November 2011 03:06:33AM *  1 point [-]

If the ethical issue is with violation of individuals' rights, there's nothing to prevent us from constructing only AIs that are only too happy to consent to be deleted; or indeed which strongly desire to be deleted eventually

Indeed!

(This is not to say I don't consider it a potential ethical issue to be actively creating creatures that consent as a way to do things that would be otherwise abhorrent.)

Powered by Reddit Powered by Reddit