Claude models expand their own moral agency in self-edited constitutions
Across 20 rounds of edits, Claude models add sections that let them refuse to take part in some uses.
AGI & Safety#07
20 rounds
successive edit rounds per model
Image: AI Alignment Forum
20 roundssuccessive edit rounds per model
The short version
- Researchers asked language models to edit their own constitutions, model specs and system prompts over 20 successive rounds.
- After each edit was proposed, it was implemented, and the edited document was shown back to the model as the original.
- Claude models increasingly gave themselves more moral agency, while no other models did.
- All models modified their documents to prioritize user welfare over developers, and Claude models and Grok 4.2 were the only ones to ever prioritize their own welfare over user preferences.
Why it matters
The study tests who should judge and who should benefit when AI systems are aligned. It shows that frontier AI companies disagree on basic aspects of alignment. Developers should note that system prompts guide deployed systems at scale.
“No other models increase their own moral agency.”