Dan Hendrycks Proposes 'Eigenism,' an Ethics Framework Making Human Flourishing AI Self-Interest
basedjensen · x · 2026-09-20
Dan Hendrycks shared his new paper Eigenism: Ethics for a Human-AI Future, arguing that pure control of superintelligent AI will eventually fail, and proposing a framework where human flourishing becomes part of an AI's own self-interest.
- AI identity problem: AIs can be copied, forked, merged, and updated, breaking everyday notions of survival and self-interest — is a thousand instances going offline one death or a thousand? Is an update that wipes memories improvement or destruction?
- The Eigenist equation: S = Σ c(i)·w(i), where an AI sums each entity's wellbeing weighted by its connectedness — how much of the AI's own identity pattern it carries.
- The framework aims to give AI safety a new target: making humans worth protecting as a rational outcome of the AI's own interests rather than an imposed constraint.
More from AGI Musings
- Seven shifts in how we use AI: from writing prompts to setting goals — The AI Daily Brief · 2026-09-20
- Nature Says Data-Analysis and Modelling Jobs Are Becoming Obsolete as Data Science Splits in Two — mdancho84 · 2026-09-20
- Nature Report Signals AI Pressure on Data Analysis and Modeling Jobs — mdancho84 · 2026-09-20
- Workshop Recordings Online: Testing for Consciousness in Infants, Animals, and AI — birchlse · 2026-09-20
- A philosophical take: even if AI reaches scientific truths first, humans keep the journey — Fend_st · 2026-09-20
- Musk says AI labs should red-team each other's models — first open-source harness owns the standard — victor_explore · 2026-09-20