---
type: "Evidence Item"
title: "Measuring political bias in Claude"
description: "We want Claude to be seen as fair and trustworthy by people across the political spectrum, and to be unbiased and even-handed in its approach to political topics. In this post."
resource: "https://www.anthropic.com/news/political-even-handedness"
tags: ["appendix-iii", "benchmark", "anthropic"]
timestamp: "2025-11-13"
category: "benchmark"
publisher: "Anthropic"
cope_score: 86
confidence: 0.88
---

# Measuring political bias in Claude

# Claim

We want Claude to be seen as fair and trustworthy by people across the political spectrum, and to be unbiased and even-handed in its approach to political topics. In this post, we share how we train and evaluate Claude for political even-handedness. We also report the results of a new, automated, open-source evaluation for political neutrality that we’ve.

# Relevance

Appendix III, section one: model and benchmark capability evidence

# Oracle Verdict

Anthropic is describing a frontier or production capability that pushes directly on the thesis. The important signal is not the marketing language; it is the widening set of tasks now being routed through model-driven execution rather than ordinary software or headcount.

# Metadata

* Publisher: Anthropic
* Category: benchmark
* Sector: General AI capability
* Capability: Model and benchmark capability movement
* Cope score: 86
* Confidence: 0.88

# Related Concepts

* [Live evidence index](index.md)
* [Thesis](../thesis.md)

# Citations

[1] [Measuring political bias in Claude](https://www.anthropic.com/news/political-even-handedness)