Thinking about maximization and corrigibility

James Payor

2023年4月21日

Published on April 21, 2023 4:03 AM GMTThere are broad issues with AI designs like "we've given you some function f(x).mjx-chtml {display: inline-block; line-height: 0; text-indent: 0; text-align: left; text-transform: none; font-style: normal; font-weight: n…

There are broad issues with AI designs like "we've given you some function f(x), now work on maximizing it as best you can". I think the issues basically boil down to Goodhart's curse. When you opti… https://www.lesswrong.com/posts/AFJgo99YckQnhbF8Z/thinking-about-maximization-and-corrigibility