Skip to content
Safety

GPT Models Write Their Own Constitution.

LessWrongAnalysis or commentary·
AI brief

Anthropic published a document called "Claude's Constitution" outlining the values and behavior intended for Claude, which Claude helped draft and edit.

Why it matters: It shows an AI model contributing to the drafting of its own guiding principles as models become more capable of cooperating with each other.

Written by AI from LessWrong's published text.

Background

From our own funding, model and news trackers, not from the source.

Source

“Earlier this year, Anthropic published "Claude's Constitution" , a vision for Claude's values and behavior. It was primarily an Anthropic project, though Claude helped draft and edit it.”
LessWrong, 27 Oct 2026, 23:17.

Interpretation or community commentary rather than straight reporting.

Read original story ↗