I think the answer to "Can AGI become aware?" is yes, but deep-learning is not the way to get there. The reason why there are very different types of neurons i

I think the answer to "Can AGI become aware?" is yes, but deep-learning is not the way to get there. The reason why there are very different types of neurons in the brain is that those neurons have different functions they perform - the total number of functions is low, but the nested effect is profound. I think the key is in understanding how we humans use concepts first, fully understanding the underlying functionality and various graphs of relations, and then building that active system in the form of an AGI.