(1) A prediction market has two jobs. The first, putting a price on the future, gets all the attention. The second only shows up at the end: deciding what the future turned out to be, so the right people get paid.
Most of the time the second job is easy. The election has a winner, the game has a score. The hard part arrives when the answer is worth fighting over, when a market grows large enough that bending the outcome pays better than betting well. At that point, whoever settles the result is effectively holding everyone's money, and every ambiguity in the question becomes a target.
Every resolution process comes under this kind of pressure once enough money shows up, and Augur has been working on the problem for over a decade.