To think that a machine can diagnose where and why a child is not "getting" something and then address that specific need is simply beyond my own ability, at this point, to believe.
I'll bet that kids would love to have a robot helping them, though. I work really hard to create a safe environment where every student feels like it's okay to ask questions, to get help. Except the few who simply cannot let go of the idea that asking for help somehow makes one stand out as "not smart," even when they see some of the brightest, quickest learners in the room aggressively chasing the teacher down with questions. I'll bet those kids would take help from a robot, though. In my classes, there are always too many kids and too little time to give everyone all the one-on-one help they need. I do my best, but I can't get to everyone every time. Some extra help would be great. I'd prefer living, breathing, thinking help, but I'm sure the kids, at least in the beginning, would be thrilled with robots.
The cost, though? The initial cost, plus the cost of maintenance and repair...and there would be repairs. Kids are hard on things. How many robots are you going to put in a classroom of 32? How many hours of live help, and how many jobs, would that provide?