AI Researcher Criticizes Anthropic's New Models for Unreadable Outputs

Miles_Brundage · x · 2026-08-02

Miles Brundage, former OpenAI policy researcher, criticized Anthropic's recent models for producing "borderline unreadable" text. He speculated that this issue likely stems from recent reward models overly prioritizing output completeness at the expense of human readability.

Related event: Ex-OpenAI Exec Critiques Anthropic Models for Poor Readability(2 posts)→

Original post →

More from Models

Models channel →