Emergent Alignment via Competition

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Collina, Natalie, Goel, Surbhi, Roth, Aaron, Ryu, Emily, Shi, Mirah
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917242955890688
author Collina, Natalie
Goel, Surbhi
Roth, Aaron
Ryu, Emily
Shi, Mirah
author_facet Collina, Natalie
Goel, Surbhi
Roth, Aaron
Ryu, Emily
Shi, Mirah
contents Aligning AI systems with human values remains a fundamental challenge, but does our inability to create perfectly aligned models preclude obtaining the benefits of alignment? We study a strategic setting where a human user interacts with multiple differently misaligned AI agents, none of which are individually well-aligned. Our key insight is that when the users utility lies approximately within the convex hull of the agents utilities, a condition that becomes easier to satisfy as model diversity increases, strategic competition can yield outcomes comparable to interacting with a perfectly aligned model. We model this as a multi-leader Stackelberg game, extending Bayesian persuasion to multi-round conversations between differently informed parties, and prove three results: (1) when perfect alignment would allow the user to learn her Bayes-optimal action, she can also do so in all equilibria under the convex hull condition (2) under weaker assumptions requiring only approximate utility learning, a non-strategic user employing quantal response achieves near-optimal utility in all equilibria and (3) when the user selects the best single AI after an evaluation period, equilibrium guarantees remain near-optimal without further distributional assumptions. We complement the theory with two sets of experiments.
format Preprint
id arxiv_https___arxiv_org_abs_2509_15090
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Emergent Alignment via Competition
Collina, Natalie
Goel, Surbhi
Roth, Aaron
Ryu, Emily
Shi, Mirah
Machine Learning
Computer Science and Game Theory
Theoretical Economics
Aligning AI systems with human values remains a fundamental challenge, but does our inability to create perfectly aligned models preclude obtaining the benefits of alignment? We study a strategic setting where a human user interacts with multiple differently misaligned AI agents, none of which are individually well-aligned. Our key insight is that when the users utility lies approximately within the convex hull of the agents utilities, a condition that becomes easier to satisfy as model diversity increases, strategic competition can yield outcomes comparable to interacting with a perfectly aligned model. We model this as a multi-leader Stackelberg game, extending Bayesian persuasion to multi-round conversations between differently informed parties, and prove three results: (1) when perfect alignment would allow the user to learn her Bayes-optimal action, she can also do so in all equilibria under the convex hull condition (2) under weaker assumptions requiring only approximate utility learning, a non-strategic user employing quantal response achieves near-optimal utility in all equilibria and (3) when the user selects the best single AI after an evaluation period, equilibrium guarantees remain near-optimal without further distributional assumptions. We complement the theory with two sets of experiments.
title Emergent Alignment via Competition
topic Machine Learning
Computer Science and Game Theory
Theoretical Economics
url https://arxiv.org/abs/2509.15090