Dietary defaults in Claude Opus 4.8

A Substack post argues that Anthropic's Claude Opus 4.8 model could be better aligned with its constitution by adjusting dietary defaults, suggesting a high-leverage improvement for AI safety.

This is a linkpost for https://open.substack.com/pub/violentlywilting/p/dietary-defaults-in-claude-opus-48?r=1txogj&utm campaign=post&utm medium=web&showWelcomeOnShare=true https://forum.effectivealtruism.org/out?url=https%3A%2F%2Fopen.substack.com%2Fpub%2Fviolentlywilting%2Fp%2Fdietary-defaults-in-claude-opus-48%3Fr%3D1txogj%26utm campaign%3Dpost%26utm medium%3Dweb%26showWelcomeOnShare%3Dtrue An easy, high leverage way Claude could be more aligned with its constitution.