Different moral theories shape superintelligence differently; Anthropic leans virtue ethics while OpenAI favors deontology

geoffreyirving · x · 2026-08-23

Different moral theories extrapolate to superintelligence in vastly different ways, necessitating a discussion on which is preferable. Modern AI character training reveals divergent philosophical bets: Anthropic leans towards virtue ethics, while OpenAI favors deontology.

Related event: Geoffrey Irving on Character Training: Language Philosophy Challenges and Diverging Moral Philosophy Bets in AI Alignment(7 posts)→

Original post →

More from AGI Musings

AGI Musings channel →