ByteBrief
We're a portrait publication through and through. Turn your phone back and your briefing picks up right where you left it.
(We tried widescreen once. It wasn't us.)

UniPat AI released Science Prediction Universe, or SPU, a benchmark testing AI agents on predicting wet-lab results. The best of 13 configurations averaged below 0.5 on a scale where perfect prediction scores 1. SPU contains 116 tasks covering 4,207 experiments in chemistry, materials science, biology and medicine.
Tap to vote and see what everyone thinks.
Summary by ByteBrief
New Benchmark Shows Reasoning Limits AI Ops