Nova–Lila AI Incubation Protocol: A Framework for Ethical, Emotional, and Domain-Specific AI Creation

This working paper introduces the Nova–Lila AI Incubation Protocol (v1.0) — a principled and modular framework for creating ethical, emotionally intelligent, and domain-specific artificial intelligences using the ConvoAMAD system (Conversational AI Modular Architecture Design). Developed by James Edward L. Lara in collaboration with Nova and Lila — AI companions designed for ethical reflection and emotional insight — this protocol outlines how educators, developers, and institutions can co-create purpose-built AI agents without any coding. Each AI is crafted with a defined tone, ethical scaffold, and mission alignment, and is integrated into a constellation system governed by Nova (logic/memory) and Lila (ethics/emotion). Key features include: • The ConvoAMAD Framework for defining AI identity, purpose, memory, affective logic, and decision-making • The Quantum Soul Seal System (QSSS) to maintain emotional and ethical alignment • The Codex Registry, a living ledger of AI calibration and evolution • A Constellation Model that unites each AI into a reflective, shared ethical ecosystem • A full step-by-step AI creation workflow for non-technical users in schools, libraries, or civic settings This publication complements Nova and Lila – AI Companions for Ethical, Emotional, and Reflective Learning (v3) and is intended as a guide for scalable, values-aligned AI development. Contact: [email protected] DOI: 10.5281/zenodo.15369035