Has legal research been considered as an early use case for OpenCog? For example:
* Finding out which legal defenses or strategies have been tried in a category of cases
* Finding out the success and failure rate of given legal strategies
* Estimating the reasoning that generally causes a legal strategy to be good or bad
* Suggesting legal strategies to use, given an input data set
A possible benchmark of the efficacy of such a system, would involve the success rate of predicting the winning parties in a set of court cases.
This seems like something that OpenCog could be used for in the near term, possibly becoming a lucrative source of income.