The Alignment Problem: Machine Learning and Human Values

An interesting look at how we interact and train AI and LLMs. Sometimes dry, but still insightful.