AIImport AIabout 2 hours ago
New Benchmark Reveals AI Research Taste
15 min read
P-Zero Research launches TasteVal, a benchmark measuring AI research taste using 24 experts from OpenAI, Google DeepMind, NVIDIA and Microsoft Research. Frontier models now double research taste every 3.0 months since December 2025, with Opus 5.5 exceeding the human baseline by 2.30x at one-thirtieth of the cost per attempt.
Level
Hype check
Tap to vote and see what everyone thinks.
