I don’t think it would be “directly” observable. Almost nothing in physics now days is “directly” observable, what ever that means.
However, if perhaps QM describes the equilibrium state of some sort of “mechanical” evolution of a system outside of view, it might take time to reach equilibrium. As entanglement is more complex, over wider distances, and probed more rapidly, it is possible that deviations from standard QM might be observed. That is the “test” I see on the horizon with quantum computers. In fact the whole endeavor of quantum computing depends on the yet untested assumption that quantum entanglement will work with out modification up to hundreds of atoms, or even more. At the moment, this is an untested assumption, which may or may not prove to be the case.
Quantum computing and communication, therefore, might be one place, for example, where some pilot wave frameworks might begin to produce different predictions, especially if it is a sub-luminal wave that takes exponential passes to resolve entanglement, perhaps does not ever find equilibrium in some cases, or perhaps has more than one possible equilibrium in some cases.
I’ve run this by some physicists and they don’t think I’m crazy. If you guys take it up and write something on it, let me know. I’d love to get a middle authorship on something like that, though I don’t have physics chops to fully develop it.