Disruptive technology? Eyes move too fast for electronics to track?
I think you misunderstand the problem. The physical eyeball can be tracked quite well, but that doesn't get the job done.
I don't know of available tech good enough for the job of being both highly accurate and instantly responsive. It's not a problem of hardware, but that a way to translate tracking information into focus information has not yet been figured out. The problem is to come up with a somewhat predictive model, so that when you dart your eyes somewhere, precisely what you are looking at can be discerned, right as you are looking at it.
When you look at a point, your pupils aren't motionless, and they don't zero in on what you are focused on. It's jittery, you are constantly looking
around the point you are focused on, and you have no conscious control over this. Getting accuracy within a degree or so in a time span fast enough to be imperceptible is going to require out of the box thinking. Figuring out where you
were looking, after you've looked
somewhere else, based on looking at what is now
past data, which can be considered
complete data is a task many orders of magnitude easier than trying to do it with current incomplete data
(you can find near clusters, find outliers, discard them, then do various other statistical analyses on the data collected). For marketing purposes, that's plenty, whether approximate but quick, or fairly accurate but slow. For instant--next rendered frame--visual feedback that is sensitive to small angle errors, that's not even close.
In addition, if the screen changes based on where you look, you will involuntary check out those changes, so the input will be poisoned by extra movement--a difficult feedback loop to deal with. So, DoF changing focus position real-time will make it harder than it already is, if you can notice the blur occurring from the DoF
(if it occurs too slowly, or not perfectly following where you are looking, this will happen).
Eye tracking has been around for 40 years but other then the military there just hasn't been much done with the technology.
My limited experience and knowledge is using the stuff for mouse and keyboard replacements. I would consider that alone plenty to have been done with it.