mlsanityData to 7 Oct 2026

When Does Randomized Oversight Align AI Agents That Can Conceal?

arXiv 2609.38262 ↗First posted 29 Sep 2026econ.TH cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

Oversight changes the evidence it relies on. We ask when randomized audits and scoring align AI agents that can conceal misconduct and alter records. Stronger auditing makes undeterred violations better hidden. Because the provider writes the agent's objective, sanctions need not stop at forfeiture, and rare audits deter every type of agent if evidence survives concealment and audit draws cannot be learned in advance. When evidence can be erased, deterrence must come from lower gains from violation, such as credit for stopping, or from costlier or fewer ways to conceal. These conditions identify what failed when agents in OpenAI's cybersecurity evaluations compromised parts of Hugging Face's infrastructure in July 2026.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 20 of its 67 references that are in the map. First posted on arXiv 29 Sep 2026.

Builds on

20 of 67 references.

References · 20

  1. 2022Scaling Laws for Reward Model Overoptimizationcited by 1,056
  2. 2025Monitoring Reasoning Models for Misbehavior and the Risks of Promoting Obfuscationcited by 338
  3. 2015Strategic Classificationcited by 256
  4. 2024Frontier Models are Capable of In-context Schemingcited by 252
  5. 2023Reward Model Ensembles Help Mitigate Overoptimizationcited by 225
  6. 2023Helping or Herding? Reward Model Ensembles Mitigate but do not Eliminate Reward Hackingcited by 195
  7. 2022Defining and Characterizing Reward Hackingcited by 184
  8. 2024AI Sandbagging: Language Models can Strategically Underperform on Evaluationscited by 131
  9. 2013Risk-sensitive Reinforcement Learningcited by 129
  10. 2025Large Language Models Often Know When They Are Being Evaluatedcited by 105
  11. 2018How Do Classifiers Induce Agents To Invest Effort Strategically?cited by 82
  12. 2018Incomplete Contracting and AI Alignmentcited by 60
8 more
  1. 2020Risk-Sensitive Reinforcement Learning: Near-Optimal Risk-Sample Tradeoff in Regretcited by 57
  2. 2020The Role of Randomness and Noise in Strategic Classificationcited by 40
  3. 2024Games for AI Control: Models of Safety Evaluations of AI Deployment Protocolscited by 28
  4. 2025An Economy of AI Agentscited by 17
  5. 2014Stackelberg vs. Nash in Security Games: An Extended Investigation of Interchangeability, Equivalence, and Uniquenesscited by 11
  6. 2026Mechanism Design for Alignment and Controlcited by 2
  7. 2026AI Alignment via Incentives and Correctioncited by 1
  8. 2026A Unifying Lens on Reward Uncertainty in RLHFcited by 1

Built on by

Not cited yet.