Learn about this first
AI safety researcher Dan Hendrycks published a post distilling the key points of his team’s new paper, “Eigenism.” He proposed 13 core principles, including the concepts that “identity and survival are graded,” “shared information determines moral obligations,” and the “auditor’s wager,” and summarized them in both two-sentence and one-sentence versions. The theory argues that, in the context of AI surpassing human intelligence, controlling AI alone is a limited strategy. Humanity must use engineering methods to build a future that incorporates its own survival into AI’s logic, achieving mutually beneficial coexistence between humans and AI.
Latest developments
Dan Hendrycks Quotes Daniel Dennett and Promotes New Paper
Event progress2
Dan Hendrycks Quotes Daniel Dennett and Promotes New Paper
Dan Hendrycks: Dan Hendrycks Quotes Daniel Dennett and Promotes His New PaperDan Hendrycks Distills the Core Ideas of the “Eigenism” Paper: A Strategy for Human Survival in the AI Era
Dan Hendrycks: Dan Hendrycks Distills the Core Ideas of the “Eigenism” Paper: A Strategy for Human Survival in the AI Era