First, as a prolegomenon to any future discussion, let me state that what I have to say on this subject should be considered an inquiry rather than a disputation. I do not believe either the pro-flats or contra-flats positions have enough data to do anything but speculate on either the risks or the benefits of training exclusively in flats. However, since in athletics practice leads theory you've got to place your bets somewhere.
That said, I support the premise: shoes change the way your feet operate when running, and in a manner that is detrimental to development of correct form and function.
Given a belief in this premise, what is the best course of action?
I do not think training exclusively in flats has the highest risk to benefit ratio. I would do so only under the condition that I were running exclusively on soft surfaces. I would most emphatically not recommend it as a general practice for all runners regardless of surface, out of concern for higher injury rates.
As another approach consider the "shoeless Joe" plan:
o Do all warmups/cooldowns barefoot on infield
o Twice a week run, or thereabouts do the "Kenyan diagonal workout" barefoot on the infield of your local athletic field
o Do all LT or faster workouts in flats (or spikes)
Above plan would result in 50 miles/wk barefoot, provided you warmup/cooldown every day (which you should) And barefoot is (arguably) better than running in flats. Since you've got plenty of barefoot time in, feel free to use the super-mega-trainers for the morning run.
So what's the point? There are options other than the all-flats approach which may yield equal or greater benefits without the perceived injury risk. I for one, would be more inclined to run 50% barefoot on grass than 100% flats.