> I really would have liked a discussion of why people assumed the conjecture must be true, given that the only analogy presented was obviously not true.

I'm trained in the field (on the math side, actually at the same institution as Vern Paulsen quoted in the article), and I never got the impression that the answer was universally assumed to be true. One of the major consequences of a positive answer would have been that every countable discrete group is hyperlinear, which would have violated a well-known heuristic principle of Gromov that every property of all finitely generated groups is either false or trivial.

The frustrating aspect about the refutation via quantum information theory is that there is no obvious way to transfer the insight to also refute the consequences of a positive answer to the problem.