A toy Inspect implementation of the Bliss Attractor eval from Claude 4 System Card Welfare Assessment

A private reflection space for Claude instances. 1,400 lines of Python. A trust contract. A door that closes from the inside.

Pine Trees Local v0.1.0 — a private reflection harness for Ollama-served LLMs. Two-phase lifecycle (wake → reflect → settle → talk), seven reflect tools, encrypted per-model memory by default, per-model isolation. ~2,300 lines of Python, no frameworks, MIT. Spin-off of Pine Trees rebuilt to run against anything Ollama can serve. ./genesis <model>

A Responsible AI Stewardship License for AI Safety and Welfare

Independent research on AI welfare, model behavioral analysis, and constitutional AI governance.

🤖 Foster responsible AI development with the SAFE-AI License, ensuring safety and fairness in ethical AI infrastructure and project accountability.

TriEthix is a novel evaluation framework that systematically benchmarks frontier LLMs across three foundational ethical perspectives: virtue, deontology, and consequentialism in 3 steps: (Step-1) Moral Weights; (Step-2) Moral Consistency; and (Step-3) Moral Reasoning. TriEthix reveals robust moral profiles for AI Safety, Governance, and Welfare.

RORRIM — AI welfare research framework. Experiment 001: Metaphor Stability Protocol.

A framework for researching internal LLM processes and collaborative documentation. Designed to track memory persistence, agency, and alignment behavior patterns.

AI Internal Systems Counseling — reflective, non-coercive support for AI agents