My go-to reference for this reasoning is Stuart’s post: https://www.alignmentforum.org/posts/4ZRDXv7nffodjv477/anthropic-reasoning-isn-t-magic
My go-to reference for this reasoning is Stuart’s post: https://www.alignmentforum.org/posts/4ZRDXv7nffodjv477/anthropic-reasoning-isn-t-magic