Benchmark · Forecasting
Future Prediction Bench
Grades AI on whether its forecasts come true.
- What it measures
- Forecasts on live MLB and USGS event feeds
- Scored by
- The event itself, vs a baseline sealed in advance
- Headline figure
- Live
- Domain
- Forecasting
- Applies to
- Forecasting and prediction markets
- Access
- On request
How it is scored
Real events
Forecasts are scored when the event happens, against a baseline sealed in advance.
What it tests
Grading by real events
- AI forecasts live events. We score it when they happen, against a baseline sealed before the event. Nothing to memorise in advance.
- Major League Baseball (MLB) and US Geological Survey (USGS) earthquake feeds are live. Every day adds data that cannot be recreated later.
Experts behind it
Practitioners define what correct means for this benchmark. Paid, remote roles.
Other benchmarks