> Unitarity says the quantum mechanical probabilities of all possible outcomes of a particle interaction must sum to one. To prove it, one would have to observe the same interaction over and over and count the frequencies of the different outcomes. Doing this to perfect accuracy would require an infinite number of observations using an infinitely large measuring apparatus, but the latter would again cause gravitational collapse into a black hole. In finite regions of the universe, unitarity can therefore only be approximately known.

I'm sure the physicists have something coherent in mind, but the discussion of unitarity in the article makes no sense at all. The idea that the sum of all probabilities of all possible outcomes of some situation is 1 isn't even part of physics; it's the definition of probability. It's not necessary to make an infinite number of observations, or even a single observation, to know that the sum over all the possibilities is 1 (also phrasable as "something will happen"). If we did make an infinite number of observations, the only possible sum over all possibilities would be 1, because it's 1 by definition. No matter what we observed, it would still be 1, and we would assign probabilities based on our observations so that the sum of them all would be 1.

The previous HN discussion linked by Hopka seems to state that in fact unitarity is not in doubt, that it's just that under the system described in the article no one has yet proved that a non-unitary solution cannot be generated. But how could unitarity possibly be "suspect"? It's a definition.