WHAT NEW IN X11.04 WED

AI collections from X · 22 collections · 66,612 posts

jevPOST

Prediction market classifiers tested: Jev vs Liquid AI D1

Using Tetlock's historical forecasting problem set, the author had Jev and D1 assign probabilities to each outcome and compared them with actual results; D1 won 9 of 12 categories.

utkarsh ruhela
@UtkarshRuhelaArchived 2026.10.03Originally posted 2026.10.02
utkarsh ruhela @UtkarshRuhelaOriginally posted 2026.10.02
Open original on X

Using Classifiers in Prediction Markets: @typesafeai AI's Jev vs @liquidai's D1. Liquid AI's classifier wins in 9 out of 12 categories.  About Benchmark: I learned about @PTetlock's @GJPForecasting 4 years ago, and ever since then, I have been fascinated with scenario analysis and the art of forecasting. In this study, I use the historical forecasting problem set. The classifiers assign a probability to each outcome in a scenario. Then, we compare the probabilities with the actual outcome, and we compare the quantitative results. These are very encouraging results for my prediction markets friends who are looking for newer quantitative techniques to trade on. But, as you know me- I don't condone gambling, so let me discourage the use of this tool in gambling :) Find the link in the comments.

x.com/utkarshruhela/status/2105836289576800326Archived 2026.10.03 10:05