To create mouse-level intelligence doesn't mean writing a program that behaves like a simulated mouse in a limited set of situations. It means creating something that can learn the same range of things (which is enormous) with the same experience efficiency and risk efficiency.

Experience efficiency refers to the conversion ratio between the space of situations you've experienced and the space of situations you've then become able to handle. Underrated concept

Risk efficiency is an even more underrated concept. It refers to the fact that, to learn something in the real world, you need to fuck around and find out, which carries risk. A more aggressive exploration strategy might result in faster learning, but might also get you killed.

To be intelligent is to optimize your existence trajectory to maximize experience while minimizing risk -- and converting this experience into learnings applicable in future situations with maximal efficiency.

The really hard part here is the "future" bit. You're optimizing for an unknown objective -- the end goal is to be able to operate in the situations you'll face *next*, which aren't like the situations you've seen before.