Having watched the presentation and read through the white paper, this implementation of decentralized dispute resolution is a terrible idea. Let me explain why.
First, and probably foremost, there is no security mechanism for determining the fitness of oracles. As per the white paper, oracles are simply JUR token holders. It says nothing about their understanding of legal principles or concepts of justice and fairness. More than that, one person or group can spawn hundreds or thousands of accounts. Since adjudications are handled by consensus, it wouldn't be a terribly expensive proposition for business or individuals to buy enough votes to guarantee their desired outcomes.
I don't know about you, but this is the opposite of justice. This is truly a system where money buys justice.
Second, even assuming the gaming issue doesn't materialize, JUR's assertion that game theory shows that users will vote for justice is totally off-base. Voters vote for an economic incentive. You cannot incentivize justice without first defining it and then offering an additional incentive for voting for just outcomes. There is no jurisprudence laid out in the JUR system, and from everything I've read, they have no intention of doing so. They want rules to be developed on the blockchain. The problem with this is that the rules that govern the judicial process (and arbitration) have been developed over centuries of jurisprudence and precedence. Unless they take the time to set down the rules at the outset, contract disputes will be resolved haphazardly without any sort of uniformity. This creates unprecedented uncertainty, which is a killer for any legal system.
Lastly, smart contracts are not the end-all, be-all. The reason that doctrines such as equitable estoppel exist is that the mere fact a contractual situation exists does not justify enforcement. Promissory estoppel is a great example of a doctrine that is totally absent from smart contracts: a party relies on a promise, without receiving any sort of valuable consideration, and performs the consideration they had offered in exchange for the promise. Courts have held for more than a two centuries that the party who made the promise, even without exchanging value for the performance by the other party, is liable for that value. Why? Because they held out the promise, and the other party relied on it. This doesn't exist in a contract, because this happens where a contract has not been formed or where specific terms are made outside of the bounds of the contract that give additional context to the terms within.
I'm heavily in favor of decentralizing dispute resolution. However, this is a system that is too easily gamed to be useful. The whitepaper discusses Hubs as being able to set their own rules for which oracles they'll rely on, but Hubs aren't required for adjudication. It also lacks the established jurisprudence and incentive necessary to ensure just outcomes, or even encourage them over unjust outcomes.