The Alignment Problem: Machine Learning and Human Values
4/5. Nonfiction about the uncomfortable intersection of AI and human values – fairness, transparency, moral uncertainty, etc. Hits the sweet spot of technical detail but not too much for me, and particularly interesting because talking in detail about how to build good AI involves talking in detail about how we build good humans. There's a lot of developmental psychology and moral philosophy here.
4/5. Nonfiction about the uncomfortable intersection of AI and human values – fairness, transparency, moral uncertainty, etc. Hits the sweet spot of technical detail but not too much for me, and particularly interesting because talking in detail about how to build good AI involves talking in detail about how we build good humans. There's a lot of developmental psychology and moral philosophy here.