Accounting for Context: Shaping Moral Credences for Value Alignment

In three linesTheoretical paper on moral alignment of AI agents. Proposes a formalized framework for agent decision-making under moral uncertainty accounting for contextual factors. Shows that the weak Pareto principle is violated in this setting, revealing limitations of aggregation mechanisms that ignore contextual factors (Simpson's paradox).

Summary generated by Claude — human-verified