F Visit openforecaster.github.io/futuresim →
Accessoss
MakerOpenForecaster (Shashwat Goel, Nikhil Chandak, Arvindh Arun, et al.; ELLIS Institute Tübingen / Max Planck Institute for Intelligent Systems / University of Stuttgart / University of Southampton)
Released2026-05-14
CategoryAI agent benchmark / forecasting evaluation framework
The lenses
Scored on its own merits — not rolled up from the finds we publish.
Novelty 3
Impact · breadth 3
Impact · depth 3
Actionable 4
Substance 5
Hype 2
Active in
Related concepts
Appears in finds
Top 5 · FutureSim
Lens
Window
Overall = breadth + depth + substance + ½·novelty + freshness (a small boost that halves each week)
01 
See the full feed → paperArtificial Intelligence
The forecast that won't update
A new benchmark feeds models real news in the order it actually arrived and asks them to keep revising their predictions — and finds they mostly cling to their first guess.
Overall▄▃▄▅▂▇medium · independentELLIS Institute Tübingen / Max Planck Institute for Intelligent Systems
No finds in this window — widen it to see more.
How this connects
Tap a node to open it