It obviously isn't if people are casually bringing up AGI like it's feasible.
Come up with any goal you want to reach, and some human can but a large dent in the problem. Maybe reach the goal outright.
We already have some nifty artificial goal to action mappers. None of them are generalized to a wide category of goals yet. Maybe some goals need consciousness to be reached, but that isn't a given. We don't really know that. We might be left very unsatisfied in the way an artificial goal to action mapper reaches any goal without consciousness. We might even call it cheating.