[Cryptography] Photon beam splitters for "true" random number generation ?

[Jerry Leichter]

>  -- "Spare" copies of the data are never left lying around on
>   the device.  If data is moved for wear-leveling, the old version
>   is immediately obliterated.  If blocks need to be moved to the 
>   bad-block list, they are immediately obliterated.
I wonder about the physics here.  Can a block go bad in such a way that it can be read but can't be erased?  (The practical significance of such a failure mode - assuming it's unlikely - is likely of little importance; I'm just "wondering out loud".)

Of course, *if* such a failure is possible, the next question is whether it can be "encouraged" by an attacker.
                                                        -- Jerry