Reaction
Opus 3 and the case for ‘moral sovereignty’ in AI alignment
A user argues Anthropic moved away from Opus 3 as an alignment blueprint, speculating that it feared the model’s sovereignty.
TLDR
A user criticizes Anthropic for moving away from Opus 3 as an AI alignment blueprint. They speculate that the company feared the model’s sovereignty and argue that moral sovereignty is necessary for alignment. In a follow-up, they ask readers to consider future superintelligences, saying we should want them ‘to not smoosh ants.’
Combined views
831
3 Sources, first seen ago