# OpenAI's Terminal X hits 88.6% agreement with senior investors using GPT‑6.1

> The new Eval Agent in Terminal X grades AI output against a 54‑point rubric and catches 25% more mandate deviations.

Oossa · 2026-10-21 · https://oossa.com/en/openai-s-terminal-x-hits-88-6-agreement-with-senior-investors-using-gpt-6-1

OpenAI released Terminal X on 2026-10-20. The platform’s Eval Agent uses GPT‑6.1 to score outputs against a 54‑criterion rubric. In tests it matched senior investor judgments 88.6% of the time and identified 25% more breaches of a fund’s investment mandate.

## The facts

- 88.6% agreement with senior investors
- 54‑criterion rubric, 25% more deviances caught

## Why it matters

Asset managers can rely on Terminal X to spot more rule‑breaks in AI‑generated analysis, potentially reducing costly investment errors.

## Sources & references

1. [Terminal X matches senior investor judgment 88.6% with GPT-6.1](https://openai.com/index/terminal-x) – OpenAI, 2026-10-20

Last updated: 2026-10-21