Needing to run in a specific fluid seems like it could pose longevity issues because a motor inherently needs a shaft to pass out of the sealed enclosure, causing ingress or egress - car AC compressors have this issue.
If you could also make the fluid an effective refrigerant, then this could be good for refrigeration compressors. Those run entirely in a sealed system anyway.
This seems to be a very high torque, low speed motor, operating at 360W 18Nm which means it's 190RPM (20rad/s).
With all the parallel plates, windage friction is going to be very high if they attempted to increase speed, which is usually the easy way to improve power density.
2kV for a fractional HP motor is really pushing things; you would need to integrate the boost converter and inverter into the motor housing. The moment you have cables above 1kVAC or 1.5kVDC, you're 'high voltage' and a raft of new rules applies.
I believe this type of motor could have very low power consumption while holding position without relying on non-backdriveable mechanics. That isn't stated in the article so I could be wrong. If that's the case, I could see this type of motor being valuable in servo/stepper type applications. Servos and especially steppers, run hot even when just holding torque since the coils have to stay energized.
A major concern would be ensuring no contamination. Any FOD, or metal filings generated from wear and tear could ruin such a motor.
That's an interesting point. It also implies much better efficiency at low speeds than most motors.
Given a few generations of better semiconductor, it could end up being very interesting for (railway mostly?) traction motors.
Low speed high torque means you don't need a further reduction gearbox.
Good performance near zero speeds mean you might not need to use braking at all aside from parking and emergencies.
High voltages are already widely used and available - 1500VDC nominal is an older standard for metro trains; 3kVDC is common both for older overhead and as an intermediate DC bus voltage for AC overhead. Future semiconductor generations could allow direct use of 25kV overhead (~40kVDC rectified at maximum line level) without the need for an intermediate bus, assuming the dielectric fluid was good enough.