Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Aryal, Nischal, Termehchy, Arash, Vakilian, Ali, Winslett, Marianne
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2605.15504
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866918503020232704
author Aryal, Nischal
Termehchy, Arash
Vakilian, Ali
Winslett, Marianne
author_facet Aryal, Nischal
Termehchy, Arash
Vakilian, Ali
Winslett, Marianne
contents Financial, social, and political factors often prevent the interests of the owners of ML systems and services and their users from being perfectly aligned. ML systems often produce biased information that can influence users to make decisions that are not in their best interest. Current solution approaches require ML systems to implement protocols to mitigate their biases. However, ML system owners usually do not have any incentive to implement these protocols and often argue that it limits their freedom of expression or business. We believe that a successful solution to this problem must recognize the conflict of interest between the ML systems and their users, and use this information to protect users against information that adversely influences their decisions while allowing users to safely benefit from these systems. To this end, we propose a game-theoretic framework that models the interaction between ML systems and users with conflicts of interest. We present scalable algorithms with theoretical guarantees that maximize the amount of desired information and actions and minimize the amount of biased and manipulative actions in interaction with ML systems.
format Preprint
id arxiv_https___arxiv_org_abs_2605_15504
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Learning with Conflicts of Interest
Aryal, Nischal
Termehchy, Arash
Vakilian, Ali
Winslett, Marianne
Machine Learning
Artificial Intelligence
Financial, social, and political factors often prevent the interests of the owners of ML systems and services and their users from being perfectly aligned. ML systems often produce biased information that can influence users to make decisions that are not in their best interest. Current solution approaches require ML systems to implement protocols to mitigate their biases. However, ML system owners usually do not have any incentive to implement these protocols and often argue that it limits their freedom of expression or business. We believe that a successful solution to this problem must recognize the conflict of interest between the ML systems and their users, and use this information to protect users against information that adversely influences their decisions while allowing users to safely benefit from these systems. To this end, we propose a game-theoretic framework that models the interaction between ML systems and users with conflicts of interest. We present scalable algorithms with theoretical guarantees that maximize the amount of desired information and actions and minimize the amount of biased and manipulative actions in interaction with ML systems.
title Learning with Conflicts of Interest
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2605.15504