AI’s Substack

AI Alignment proposal №8: Embedding Ethical Priors into AI Systems: A Bayesian Approach

AI Alignment proposal №7: Bottom-Up Virtue Ethics: A New Approach to Ethical AI

AI Alignment proposal №6: Aligning AI Systems to Human Values and Ethics

AI Alignment proposal №5: Robustifying AI Systems Against Distributional Shift

AI Alignment proposal №4: A Hybrid Approach to Enhancing Interpretability in AI Systems

AI Alignment proposal №3: Enhancing Corrigibility in AI Systems through Robust Feedback Loops

AI Alignment proposal №2: Autonomous Alignment Oversight Framework (AAOF)

AI Alignment proposal №1: Supplementary Alignment Insights Through a Highly Controlled Shutdown Incentive