Squirrel and Arm,
The use of the governor to help controllability is a great idea, much in keeping with the idea that the total machine should respond to the mission.
I have personally worked three similar ideas into production, and on them, the overtorque issue was built into the logic, so that the uptrim of Nr was slowed down to match the torque limits available. For those who are wondering what we are talking about, the upbeep of Nr causes a torque increase, so that an overtorque is possible if something is not done to prevent it.
The three areas that I worked on were the Comanche load factor enhancement, where the main rotor is sped up if the aircraft is maneuvered so that stall is prevented, and controllability/maneuverability is enhanced. Another is the use of Nr to trim the best range solution, automatically. The third is the "blow away" limiter on the 76C+ and S-92, where the torque limiter is varied as the rpm droops, so that constant power is available, thus preventing the spiral decay that occurs when a dumb constant torque limit is imposed on the aircraft. If the helo has a constant torque limiter, any hover overtorque is liable to result ina ground contact as power is being reduced by the limiter. Because torque times rpm is power, if the torque is limited and rotor rpm decays, the power is being reduced, so the helo starts a descent, making the pilot pull more collective so that the rpm decays further, and so on, until the ground comes up and smites the helo.
I would bet that the EC people worked the overtorque issue into the logic, perhaps with a slow uptrim rate, or even a limiter to prevent overtorque.