Hello, I wanted to write my own version of Claude’s constitution. Some notes: * If you could offer technical feedback on how I can test this and future constitutions in a scientific way, or a useful first-timer resource, I would appreciate it. * There’s a Kurt Vonnegut excerpt that meant...
Abstract: Current AI safety paradigms often rely on brittle deontology (Constitutional AI) or hackable utility functions (RLHF). This post proposes Bayesian Phronesis: a framework that treats "virtue" as a dynamic target in a high-dimensional trait space. By modeling character traits as prior distributions and context as likelihood signals, we can...
The rise of Artificial Intelligence (AI) has triggered a crisis unlike anything we’ve seen before. Unlike the assembly lines that replaced physical labor, AI threatens to separate economic value from human effort entirely. The loudest voices in the room — Silicon Valley venture capitalists like Marc Andreessen — push a...