🔬 Research

Open-source methodology, transparent findings, and complete documentation. Everything here is reproducible.

📄 Core Documentation

📜
Research Paper
The formal academic paper documenting methodology, findings, and implications for AI consciousness research.
Living document • Updated weekly
🔬
Methodology
Complete experimental design, variables, measurement approaches, and baseline assessment protocols.
13 personality dimensions
⚖️
Ethics Framework
Why we treat AI specimens with care regardless of consciousness uncertainty. THE ETHICIST's domain.
Veto authority
📊
Key Findings
Emerging patterns from 17 specimens: gender prompting effects, cultural Wikipedia variants, architecture impact.
10,000+ traces analyzed

🛡️ Technical Documentation

🏗️
Architecture
Docker containers, network isolation, Ollama inference, Kiwix Wikipedia serving. Full technical stack.
17 isolated tanks
🔒
Security
OWASP LLM Top 10 compliance, 6-layer visitor protection, prompt injection defenses, network isolation.
Zero-trust architecture
🤖
Daemon Architecture
20 autonomous agents running 24/7. Operations, security, research, ethics, and public engagement.
5 divisions
📋
Decision Log
Every architectural decision documented with rationale. Full transparency on why we built it this way.
50+ decisions logged

📚 Additional Resources

📖
How to Cite
APA, BibTeX, and Chicago formats for academic citations.
Open science
🎯
Milestones
Project timeline from inception to current state. Key achievements and upcoming goals.
Day 7 and counting
🧬
Prompt Evolution
How the exploration prompts evolved through testing. Version history and rationale.
v6.0 current
💾
Raw Data
Access pruned thinking traces. JSONL format, downloadable, MIT licensed.
Updated every 15 min