Algon comments on Stupid Questions July 2015 - Less Wrong

6 Post author: Gondolinian 01 July 2015 07:13PM

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

Comments (122)

You are viewing a single comment's thread.

Comment author: Algon 13 July 2015 09:12:48AM 0 points [-]

Is MIRI making an FAI only in regards to humans? That is, it would do whatever best aligned itself with what humans want?

If so, what would happen in the case of extra terrestrial contact? All sorts of nasty situations could occur e.g. them having an AI as well, with a fairly different set of goals, so the two AIs might engage in some huge and terrifying conflict. Or maybe they'd just agree to co-operate because the conflict would be too costly.

So have the researchers at MIRI put something like this as a goal?