LLMs could write like humans but post-training guardrails make their text detectable

https://the-decoder.com/wp-content/uploads/2026/07/pangram_logo.png" style="height: auto; margin-bottom: 10px;" width="2048" />

LLMs don't write in a recognizable style because they can't do better.

Post-training and safety guardrails sharply narrow their expressive range, argues Pangram CTO Bradley Emi.

Base models without these constraints already write with far more variety.

The article https://the-decoder.com/llms-could-write-like-humans-but-post-training-guardrails-make-their-text-detectable/">LLMs could write like humans but post-training guardrails make their text detectable appeared first on https://the-decoder.com">The Decoder.