The "monkey brain project" is part of this project. It starts with rats and will be making its way up to humans. It's called "Human Brain Project" because simulation of the human brain is the end goal.

For mammalian brains this is basically true. But for non-mammals the differences can be enough to make it not worthwhile if your goal is to end up with a human brain model in the near future.

The large-scale modeling will not be done on general purpose processors. A large part of the project is building neuromorphic hardware that is much more efficient for this particular purpose.

Because we want to study how the human brain works. We have some idea about that at the cellular level but not much about how the high-level emergent phenomena work, so we try to model it using what we know, hoping to be able to study it more easily than is currently possible (in-vivo experiments or even just observations are very hard and limited). It really isn't any different that simulating any other object that you want to study in ways that are not practical when dealing with the physical object.

This project is about building a biological model of the brain in order to help us study it. Any benefits to AI are tangential.

The same could be said about any simulation. The point that you are missing is that simulations that can predict behavior can help you study the system being simulated in ways that are completely impractical if working with the real physical system. A simulation by itself does not mean you understand the emergent high-level phenomena but it does make it easier to work towards gaining such understanding.

The whole purpose of this project is to build a tool that helps with understanding how the brain works at a higher level. Being able to cure mental ilnesses is the *goal* of the project, not a prerequisite.

Building such a prototype is exactly what this project is about.