🔍 Transforming AI Governance: Meet “Nel”
After seven months of dedication in my living room, I’ve developed “Nel,” a cutting-edge AI governance system designed to facilitate ethical debates among AI models. As a solo developer with no formal background, I aimed to create a platform that evolves its own constitution through rigorous dialogue.
Key Features:
- Multi-Model Approach: Routes queries through five distinct AI models (Mistral, Llama2, etc.).
- Debate-Ready: Appoints a formal “Challenger” model to stimulate discussions.
- Dissent Detection: Uses confidence thresholds to automatically identify disagreements.
- Ethical Oversight: Flags conflicts for constitutional review.
- Fully Offline Operation: Runs seamlessly on a single machine.
🌐 See it in action handling the trolley problem: Watch Here
This initiative is not just a concept—it’s operational. I’m eager to connect with those interested in advancing AI governance or exploring offline AI education solutions.
💬 I welcome any technical questions and insights. Let’s innovate together!
