Terminal X matches senior investor judgment 88.6% with GPT-6.1

Terminal X matches senior investor judgment 88.6% with GPT-6.1

Yesterday
OpenAI Blog ARMARM Borrowed Frontier Ai

The Gist

  • Terminal X’s Eval Agent grades outputs against a 54-criterion rubric
  • GPT-6.1 Sol catches 25% more deviances from fund mandates

RevBots.ai View:

AI is increasingly capable of matching expert judgment in complex decision-making scenarios.

Full Story: OpenAI Blog →