We're currently working on:

Can we solve difficult problems by assembling small and mostly context-free contributions from individual agents who don't know the big picture?

We have previously worked on:

Can we make well-calibrated predictions about human judgments after deliberation based on cheap signals such as multiple people's quick guesses?

In the long run, we might work on:

Eventually, we'll have a publicly available web and mobile app that will help you think through difficult questions using dialog. This app will integrate contributions from AI and human participants.

We're just getting started, and will add to and re-organize these projects as we grow our team.