All safety considerations aside...
Have you done any performance testing as compared with a standard kitbot drivetrain? How about one with some basic traction control?
I'm not overwhelmed by your performance videos, and I'm curious why your team made the design decision to go this direction. Was there testing or prototyping involved? All about theoretical calculations? None of the above?
If the decision was all about "cool factor" then... rock on.
However, if you believe this has higher performance than using those CIMs in the traditional "rubber meets the road" way, I'm curious why you believe so, and if you have supporting data. If you DO have supporting data, I'd absolutely LOVE to get a peek.
Enlighten a man who sometimes has difficulty understanding why others stray outside the box, when the box appears to be an optimized and elegant solution.
-John