English
Related papers

Related papers: Consequences of Misaligned AI

200 papers

In repeated games, such as auctions, players rely on autonomous learning agents to choose their actions. We study settings in which players have their agents make monetary transfers to other agents during play at their own expense, in order…

Computer Science and Game Theory · Computer Science 2026-02-12 Yoav Kolumbus , Joe Halpern , Éva Tardos

Learning to control an environment without hand-crafted rewards or expert data remains challenging and is at the frontier of reinforcement learning research. We present an unsupervised learning algorithm to train agents to achieve…

Machine Learning · Computer Science 2018-11-29 David Warde-Farley , Tom Van de Wiele , Tejas Kulkarni , Catalin Ionescu , Steven Hansen , Volodymyr Mnih

A principal designs an algorithm that generates a publicly observable prediction of a binary state. She must decide whether to act directly based on the prediction or to delegate the decision to an agent with private information but…

Theoretical Economics · Economics 2024-02-22 Ruqing Xu

Value alignment has emerged in recent years as a basic principle to produce beneficial and mindful Artificial Intelligence systems. It mainly states that autonomous entities should behave in a way that is aligned with our human values. In…

Multiagent Systems · Computer Science 2021-06-28 Nieves Montes , Carles Sierra

This paper studies a dynamic screening model in which a principal hires an agent with limited liability. The agent's private cost of working is an i.i.d. draw from a continuous distribution. His working status is publicly observable. The…

Theoretical Economics · Economics 2025-11-26 Yijun Liu

With recent development of artificial intelligence, it is more common to adopt AI agents in economic activities. This paper explores the economic actions of agents, including human agents and AI agents, in an economic game of trading…

Theoretical Economics · Economics 2026-03-03 Huan Cai , Ziqing Lu , Catherine Xu , Weiyu Xu , Jie Zheng

We study the classic principal-agent model when the signal observed by the principal is chosen by the agent. We fully characterize the optimal information structure from an agent's perspective in a general moral hazard setting with limited…

Theoretical Economics · Economics 2023-07-25 Majid Mahzoon , Ali Shourideh , Ariel Zetlin-Jones

We study partially observable assistance games (POAGs), a model of the human-AI value alignment problem which allows the human and the AI assistant to have partial observations. Motivated by concerns of AI deception, we study a…

Artificial Intelligence · Computer Science 2025-08-12 Scott Emmons , Caspar Oesterheld , Vincent Conitzer , Stuart Russell

Inferring a person's goal from their behavior is an important problem in applications of AI (e.g. automated assistants, recommender systems). The workhorse model for this task is the rational actor model - this amounts to assuming that…

Machine Learning · Computer Science 2019-03-15 Alexander Peysakhovich

Online platforms in the Internet Economy commonly incorporate recommender systems that recommend products (or "arms") to users (or "agents"). A key challenge in this domain arises from myopic agents who are naturally incentivized to exploit…

Information Retrieval · Computer Science 2024-06-19 Xiaowu Dai , Wenlu Xu , Yuan Qi , Michael I. Jordan

Reward hacking--where agents exploit flaws in imperfect reward functions rather than performing tasks as intended--poses risks for AI alignment. Reward hacking has been observed in real training runs, with coding agents learning to…

Artificial Intelligence · Computer Science 2025-08-26 Mia Taylor , James Chua , Jan Betley , Johannes Treutlein , Owain Evans

Advanced reasoning models with agentic capabilities (AI agents) are deployed to interact with humans and to solve sequential decision-making problems under (approximate) utility functions and internal models. When such problems have…

Artificial Intelligence · Computer Science 2025-09-25 Daniel Jarne Ornia , Nicholas Bishop , Joel Dyer , Wei-Chen Lee , Ani Calinescu , Doyne Farmer , Michael Wooldridge

A principal provides nondiscriminatory incentives for independent and identical agents. The principal cannot observe the agents' actions, nor does she know the entire set of actions available to them. It is shown, very generally, that any…

Theoretical Economics · Economics 2024-01-31 Ashwin Kambhampati

In the future, artificial learning agents are likely to become increasingly widespread in our society. They will interact with both other learning agents and humans in a variety of complex settings including social dilemmas. We argue that…

Artificial Intelligence · Computer Science 2022-02-22 Tobias Baumann

We study incentive design when multiple principals simultaneously design mechanisms for their respective teams in environments with strategic spillovers. In this environment, each principal's set of incentive-compatible mechanisms--those…

Theoretical Economics · Economics 2026-05-11 Brian Roberson

This report examines what I see as the core argument for concern about existential risk from misaligned artificial intelligence. I proceed in two stages. First, I lay out a backdrop picture that informs such concern. On this picture,…

Computers and Society · Computer Science 2024-08-14 Joseph Carlsmith

Human-designed reward functions for reinforcement learning (RL) agents are frequently misaligned with the humans' true, unobservable objectives, and thus act only as proxies. Optimizing for a misspecified proxy reward function often induces…

Artificial Intelligence · Computer Science 2026-01-30 Stephane Hatgis-Kessell , Logan Mondal Bhamidipaty , Emma Brunskill

Many settings of interest involving humans and machines -- from virtual personal assistants to autonomous vehicles -- can naturally be modelled as principals (humans) delegating to agents (machines), which then interact with each other on…

Computer Science and Game Theory · Computer Science 2024-08-07 Oliver Sourbut , Lewis Hammond , Harriet Wood

In practice, most mechanisms for selling, buying, matching, voting, and so on are not incentive compatible. We present techniques for estimating how far a mechanism is from incentive compatible. Given samples from the agents' type…

Computer Science and Game Theory · Computer Science 2023-12-12 Maria-Florina Balcan , Tuomas Sandholm , Ellen Vitercik

Reinforcement Learning views the maximization of rewards and avoidance of punishments as central to explaining goal-directed behavior. However, over a life, organisms will need to learn about many different aspects of the world's structure:…

Artificial Intelligence · Computer Science 2023-11-16 Thomas J. Ringstrom