Patenting quantum indeterminacy isn't possible. Patenting a machine that makes it possible to exploit that phenomenon is.
Every machine is reducible to mere physics. That doesn't mean there aren't inventive steps to make them practical.
I am mostly complaining that IMO a patent should be so narrow as to be hardware design ie copyright and as such it is much much harder to get broad applicability - which instantly reduces the ability to effectively patent principles
Lets take a well known example - the use of a cyclone induced to create a vacuum used to suck up dust (Dyson) - he could not patent that it is too broad so he had to patent exact angles of fan to cylinder and so forth - which were eventually gotten around via different manufacturers - but he held the patents on the well known means of generating cyclones, for vacuums.
I am prepared to bet that the patent was not granted for the concept of a random number generator based on quantum phenomena. You can't patent ideas, you actually need to do some engineering with at least one inventive step.
Experiments in a lab are not manufacturable, reliable, productionised products. They're just not.
This might be what you are looking for http://www.google.com/patents/US20100332285
Even so things like non-obvious are a hard hurdle - most physicists know of the experiment where you fire a single proton at a half silvered mirror and it will be quantum-indeterminate whether it bounces back - that is truly random. There is /was a patent on using that for generating random numbers and is the basis of most every poker site in the world.
It's getting to the point where patenting the business model of patent trolling is going to be the only solution
Edit Having read the GC patent I can see there is at least one software patent that might just pass muster - but really I struggle with the idea this is an invention.
Edit2 - just to clarify - look at the patent for Edisons lightbulb (ignoring the fog of war surrounding patent claims). At the time the generation of light and heat from passing a current through a wire was well known - but edisons defensible patent is a specific description of how to build his lightbulb - vacuum, alloys, resistance etc. To my non legal mind that is not a set of scientific principles, but a copyright in hardware form.