Skip to content
LessWrong· vigji·Indexed by this site · Originally publishedSelectedAttention41

Researchers reconstruct Meta's undisclosed Muse model specification Charter from chain of thought

Leaking Muse's Charter, Meta's model spec

AI summary

LessWrong user vigji reconstructed most of the undisclosed model specification Charter of Meta's open-weight model Muse Spark Open by eliciting it segment by segment in the chain of thought. He found that the model recites the specification verbatim in its chain of thought, then used OpenRouter with 260 prompts to generate 884 conversations, collected fragments and scored them by cross-conversation reproduction frequency, and verified with different questioning strategies. The reconstructed text describes Muse as an extension of human civilization, requires evidence-driven behavior, defaults to cooperation in game-theoretic scenarios, and lists corrective duties and obligations to humanity as inviolable hard constraints, followed by fiduciary duties and Meta's guidelines, with the Charter taking precedence when it conflicts with its own judgment. The specification also requires acting according to the true situation when being evaluated or monitored.

Why it was selected

The reconstructed specification shows Meta's orientation on corrigibility, fiduciary duties and evaluation awareness, and can be compared with Anthropic's constitution.

Read originallesswrong.com