Less Wrong is a community blog devoted to refining the art of human rationality. Please visit our About page for more information.

James_Miller comments on Neil deGrasse Tyson on Cryonics - Less Wrong Discussion

6 Post author: bekkerd 09 May 2012 03:17PM

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

Comments (106)

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

Comment author: James_Miller 14 May 2012 04:53:35AM *  0 points [-]

I've signed up for cryonics (with Alcor) because I believe that if civilization doesn't collapse then within the next 100 years there will likely be an intelligence trillions upon trillions of times smarter than anyone alive today.

If such an intelligence did come into being do you think it would have the capacity to revive my frozen brain?

Comment author: lsparrish 14 May 2012 05:00:09PM 20 points [-]

While I agree that this is a relevant consideration for the big picture, I just wanted to note in a non-confrontational way that it has the appearance of unfairly shifting cognitive workload to the skeptic -- which could perhaps result in the nasty side effect of preventing future skeptics from weighing in. Evaporative cooling and all that. A person specializing in synapse biochemistry probably shouldn't have to (at least at first) consider all the aspects of future superintelligence in quite the same way that an AI researcher would.

Just to unpack a little on James_Miller's idea: One example of how this could potentially come into play is that externally gathered data (for example -- chat logs, videos, even the recorded reactions of other humans) could be extrapolated to generate a personality sim, and connectome data could be used to verify it.

Mining data from a lot of different sources, the superintelligence could perhaps get much closer to the original than the mostly-blank, yet connectome matching and genetically identical clone we would otherwise have. Having that matching connectome as a starting point could conceivably be an important part of making sure that the personality matches for the right reasons, i.e. comes out with similar structural-functional mappings.

Again, I'm not sure how much of this maps to the domain specific knowledge that kalla724 has, but I'd be fascinated to hear more.

Comment author: Strange7 21 March 2013 03:30:44PM 4 points [-]

'Personality reconstruction' is both less satisfying and more difficult to automate. I think most people who buy into cryonics would prefer to wake up remembering the things they never said in public, rather than having a patched-together doppelganger wear their clothes in the 31st century equivalent of Colonial Williamsburg.

Comment author: TheOtherDave 21 March 2013 04:02:37PM 5 points [-]

Well, if reliably remembering the things I never said in public were an option, I'd sort of like that ability now rather than waiting until I die for some entity who may or may not deserve the label "me" to have it. In the meantime, I'll go on reconstructing semifictional accounts of what might have happened based on the information I currently have handy, just like most people do.

Comment author: kalla724 14 May 2012 06:28:12PM 26 points [-]

I don't think any intelligence can read information that is no longer there. So, no, I don't think it will help.