The primary error in a lot of what I've read in comments above is that the sway bars do not ever become disconnected or act as though they're disconnected. There are two independent hydraulic circuits between the two cylinders - one front one rear. Each circuit has a gas charged small expansion chamber called an accumulator. Each circuit also has a valve that opens or closes the expansion chamber from the rest of the circuit. The fluid transfer between the cylinders is 1:1 in either circuit. The result is that in "closed" position there is no dampening and it's a very rigid connection front/rear, but the two cylinders can and do still move - they do so with a very rigid master/slave function.
When the valves are in the "open" position the two circuits are dampened by the accumulators. So the master/slave movement is not fully rigid. It is not fully free either. And that's the main point I'm trying to correct. They act as a softer master/slave, not really like a disconnected sway bar. And they only have a limited range of travel. Combined with the accumulator having a gas compression spring rate (highly progressive) the result is that they have relatively soft feel near the neutral position in the travel range of the cylinders and become progressively more rigid near the ends of the piston travel, and then very rigid beyond the max travel of the cylinders. There is also a significant force applied to the body to retain it's position balanced between the front and rear axle angles. So there's a body stabilization function occurring along with a forced articulation function (the two are really the opposing reactions of the same forces).
The accumulators start with IRC 500psi charge. They're not really disconnected or fully connected. It's somewhere in between. I've often been curious how the ride quality would be with a sway bar spring rate similar to an SR5 but mounted in the KDSS system. I think that could offer a feel much closer to a true disconnected sway bar.
The 12mph is taken from this diagram I believe. I'm not sure if it's totally correct though. My reading of the diagram is that over 12mph it will close the valves if the steering rate of change exceeds the limit (you turn the wheel fast) or if the lateral acceleration exceeds a certain amount - you go around a corner fast. I'm not sure if they otherwise remain open or not - for example going 40mph in a straight section of road.
I am also unaware of anyone having played with the gas pressure to see how it might work with 100psi for example.