mlsanityData to 14 Oct 2026

Propensity Inference: Environmental Contributors to LLM Behaviour

arXiv 2604.21098 ↗First posted 22 Apr 2026cs.AI cs.CLcited by 7

Not on the map. Why?

Abstract (arXiv)

Motivated by loss of control risks from misaligned AI systems, we develop and apply methods for measuring language models' propensity for unsanctioned behaviour. We contribute three methodological improvements: analysing effects of changes to environmental factors on behaviour, quantifying effect sizes via Bayesian generalised linear models, and taking explicit measures against circular analysis. We apply the methodology to measure the effects of 12 environmental factors (6 strategic in nature, 6 non-strategic) and thus the extent to which behaviour is explained by strategic aspects of the environment, a question relevant to risks from misalignment. Across 23 language models and 11 evaluation environments, we find approximately equal contributions from strategic and non-strategic factors for explaining behaviour, do not find strategic factors becoming more or less influential as capabilities improve, and find some evidence for a trend for increased sensitivity to goal conflicts. Finally, we highlight a key direction for future propensity research: the development of theoretical frameworks and cognitive models of AI decision-making into empirically testable forms.

Key facts. Cited by 7 papers in the map (data through 14 Oct 2026). Builds on 24 of its 40 references that are in the map. First posted on arXiv 22 Apr 2026.

Builds on

24 of 40 references.

References · 24

  1. 2024Alignment faking in large language modelscited by 360
  2. 2024Frontier Models are Capable of In-context Schemingcited by 287
  3. 2019Risks from Learned Optimization in Advanced Machine Learning Systemscited by 258
  4. 2022The Alignment Problem from a Deep Learning Perspectivecited by 258
  5. 2025Emergent Misalignment: Narrow finetuning can produce broadly misaligned LLMscited by 172
  6. 2025Agentic Misalignment: How LLMs Could Be Insider Threatscited by 158
  7. 2025Measuring AI Ability to Complete Long Software Taskscited by 156
  8. 2023Large Language Models can Strategically Deceive their Users when Put Under Pressurecited by 146
  9. 2025Natural Emergent Misalignment from Reward Hacking in Production RLcited by 136
  10. 2023Scheming AIs: Will AIs fake alignment during training in order to get power?cited by 97
  11. 2025Stress Testing Deliberative Alignment for Anti-Scheming Trainingcited by 86
  12. 2025Superintelligent Agents Pose Catastrophic Risks: Can Scientist AI Offer a Safer Path?cited by 75
12 more
  1. 2025Gradual Disempowerment: Systemic Existential Risks from Incremental AI Developmentcited by 46
  2. 2024Uncovering Deceptive Tendencies in Language Models: A Simulated Company AI Assistantcited by 38
  3. 2025A Rosetta Stone for AI Benchmarkscited by 36
  4. 2025Incomplete Tasks Induce Shutdown Resistance in Some Frontier LLMscited by 22
  5. 2025PropensityBench: Evaluating Latent Safety Risks in Large Language Models via an Agentic Approachcited by 22
  6. 2025Why Do Some Language Models Fake Alignment While Others Don't?cited by 20
  7. 2025Lessons from a Chimp: AI "Scheming" and the Quest for Ape Languagecited by 20
  8. 2024Planning in a recurrent neural network that plays Sokobancited by 18
  9. 2025HiBayES: A Hierarchical Bayesian Modeling Framework for AI Evaluation Statisticscited by 18
  10. 2025Evaluating the Goal-Directedness of Large Language Modelscited by 16
  11. 2025The Loss of Control Playbook: Degrees, Dynamics, and Preparednesscited by 5
  12. 2025Why Do Language Model Agents Whistleblow?cited by 5

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 7.

7 papers.

  1. 2026Multi-Agent AI Control: Distributed Attacks Hamper Per-Instance Monitorscited by 5
  2. 2026Gram: Assessing sabotage propensities via automated alignment auditingcited by 4
  3. 2026Realistic honeypot evaluations for scheming propensitycited by 1
  4. 2026LLMs Can Leak Training Data But Do They Want To? A Propensity-Aware Evaluation of Memorization in LLMscited by 1
  5. 2026Model Forensics: Investigating Whether Concerning Behavior Reflects Misalignmentcited by 1
  6. 2026Causal Behavioral Evaluation of AI Agents at Scale via Automated Behavioral Sciencecited by 0
  7. 2026AI Persuasion as a Threat to Human Controlcited by 0