8022_citations--14-.jpg

David Krueger

AI Alignment Existential Safety (x-risk) Goal Misgeneralization Reward Gaming Large Language Models (LLMs)

About

Dr. David Scott Krueger is an Assistant Professor in Robust, Reasoning, and Responsible AI in the Department of Computer Science and Operations Research (DIRO) at the Université de Montréal and a Core Academic Member of Mila (Quebec AI Institute). He holds a Canada CIFAR AI Chair and is the recipient of the IVADO Professorship in Responsible AI. Dr. Krueger is a leading voice in AI Alignment and Existential Safety (x-risk), focusing his research on ensuring that advanced AI systems—particularly Large Language Models (LLMs) and autonomous agents—act in accordance with human intent. He is widely recognized for his work on identifying "alignment failure modes," such as Goal Misgeneralization and Reward Gaming, where AI systems achieve specified rewards by pursuing unintended or harmful behaviors.

Research Performance Summary

102
Total Papers
15913
Total Citations
35
H-Index
2015-2026
Active Research Span

First Recorded Paper

Nice: Non-linear independent components estimation

Year: 2015

Citations: 3312

Venue: Workshop at ICLR

Latest Recorded Paper

Position: Capability Control Should be a Separate Goal From Alignment

Year: 2026

Citations: 0

Venue: arXiv preprint arXiv:2602.05164

Last 10 Years Publication Activity

This timeline shows the professor's yearly publication activity.

2026
3
2025
27
2024
31
2023
13
2022
7
2021
4
2020
3
2018
2
2017
6
2016
3
2015
3

Publication Venues and Collaboration

Journal, Conference, and Book Publication Breakdown

Conference 49
Journal 6
Book 1

Top Coauthors

Krueger Krasheninnikov

Research Impact by Period

2025-2026

Period Stats
Papers 30
Citations 372
Avg. Citations / Paper 12.4
H-Index 9

2020-2024

Period Stats
Papers 58
Citations 7151
Avg. Citations / Paper 123.3
H-Index 26

2015-2019

Period Stats
Papers 14
Citations 8390
Avg. Citations / Paper 599.3
H-Index 11

Contact and Professional Links

Contact Information

david.krueger@umontreal.ca

Detected Research Keywords

Out Distribution Generalization Across Language Models Learned Reward Functions Training Order Recency Recurrent Neural Networks Distributional Training Data Training Data Attribution Probabilistic Modelling Sufficient Modelling Sufficient Causal Sufficient Causal Inference