search

quantified-uncertainty / potential-projects

2 stars 0 forks source link

Forecasting: Future of Forecasting #7

Open OAGr opened 2 years ago

OAGr commented 2 years ago

We could really use at least one large question set on what the future of forecasting will look like. Some ideas:

  1. How popular will [judgemental forecasting] be in [5,10,20,50] years?
  2. How important will [monetary prediction markets], [prediction tournaments] be?
  3. How important will [public prediction markets+tournaments] vs [organization internal prediction markets+tournaments]?
  4. What domains will have them be the most useful?
  5. How much of a bottleneck/win will the following represent?
    • Long-term forecasting
    • Prediction functions, like Squiggle
    • Large-scale evaluations
    • Mostly-automated systems
  6. How significant are the downsides of forecasting? (Like, AGI companies using it for acceleration)
uvafan commented 2 years ago

Same thought here as https://github.com/QURIresearch/potential-projects/issues/15