The Conscience Layer Prototype, created by Aleksandar Rodić in 2025, establishes a research foundation for ethical artificial intelligence. It brings moral awareness into computation through principles of truth, human autonomy, and societal responsibility, defining a transparent and accountable form of intelligence.

AXON: Agentic [AI] eXploratory + Observational Network

Ethical Decision Simulator: audit how LLMs reason through moral dilemmas—multi-model compare, ethical lens stress tests, semantic agreement metrics, and exportable JSON verdicts. Python, Streamlit, Groq or Ollama—deployed on Streamlit Cloud.

Teacher-guided prompt-shape discovery for auditable moral attention in frozen weak classifiers.

Multilingual Story Moral AI Leaderboard

A mathematical framework for auditing whether LLMs execute or rationalize moral rules under irreversibility constraints.