Some mistakes in thinking about AGI evolution and control

By Remmelt @ 2025-08-01T08:08 (+7)

tl;dr A short list of mistakes made by researchers considering evolution and AGI control. They're in a long post, but hard to pick out. So I put them into bullets.
 

1. Conflating run-time environments

      (read more)

2. Simplifying the selection process

      (read more)

3. Equivocating instances of evolution

      (read more)

4. Simplifying control to constraining goals

      (read more)