Back to today’s briefing

Story perspectives

Study Reveals Left-Leaning Bias in Major AI Models

5/23/2025

49 6

Summary

Story summary

A team of researchers, spearheaded by Andrew Hall, assessed the political bias of 24 large language models from major companies like OpenAI and Google. Their findings revealed a predominant left-leaning tendency, particularly in OpenAI's responses. Hall urges AI firms to strike a balance between neutrality and user values to prevent the entrenchment of biases and echo chambers in an evolving political landscape.

1 of 1

Story summary

A team of researchers, spearheaded by Andrew Hall, assessed the political bias of 24 large language models from major companies like OpenAI and Google. Their findings revealed a predominant left-leaning tendency, particularly in OpenAI's responses. Hall urges AI firms to strike a balance between neutrality and user values to prevent the entrenchment of biases and echo chambers in an evolving political landscape.