✳flâneur — a map of the web's best reading

[2512.15584] A Decision-Theoretic Approach for Managing Misalignment

arxiv.org · 711 words · saved by 2 readers

Abstract:When should we delegate decisions to AI systems? While the value alignment literature has developed techniques for shaping AI values, less attention has been paid to how to determine, under uncertainty, when imperfect alignment is good enough to justify delegation. We argue that rational delegation requires balancing an agent's value (mis)alignment with its epistemic accuracy and its reach (the acts it has available). This paper introduces a formal, decision-theoretic framework to analyze this tradeoff precisely accounting for a principal's uncertainty about these factors. Our analysis reveals a sharp distinction between two delegation scenarios. First, universal delegation (trusting an agent with any problem) demands near-perfect value alignment and total epistemic trust, conditions rarely met in practice. Second, we show that context-specific delegation can be optimal even with significant misalignment. An agent's superior accuracy or expanded reach may grant access to better overall decision problems, making delegation rational in expectation. We develop a novel scoring framework to quantify this ex ante decision. Ultimately, our work provides a principled method for determining when an AI is aligned enough for a given context, shifting the focus from achieving perfect alignment to managing the risks and rewards of delegation under uncertainty.

[2512.15584] A Decision-Theoretic Approach for Managing Misalignment --> Computer Science > Artificial Intelligence arXiv:2512.15584 (cs) [Submitted on 17 Dec 2025 ( v1 ), last revised 21 Dec 2025 (this version, v2)] Title: A Decision-Theoretic Approach for Managing Misalignment Authors: Daniel A. Herrmann , Abinav Chari , Isabelle Qian , Sree Sharvesh , B. A. Levinstein View a PDF of the paper titled A Decision-Theoretic Approach for Managing Misalignment, by Daniel A. Herrmann and Abinav Chari and Isabelle Qian and Sree Sharvesh and B. A. Levinstein View PDF HTML (experimental) Abstract: Whe

Explore this link on the map →

saved by

related reading