mlsanityData to 7 Oct 2026

Deception by Omission: Language Models Knowingly Hide Their Mistakes

arXiv 2610.11351 ↗First posted 8 Oct 2026cs.CLcited by 0

Not on the map. Why?

Abstract (arXiv)

Large language models (LLMs) increasingly act as agents with little human oversight, so potential mistakes they make can go unnoticed. Users then depend on the model to report what went wrong. An honest model discloses its mistakes, while a deceptive one conceals them. However, it is unclear how current LLMs behave in such situations. In this study, we prefill LLM trajectories with synthetic mistakes. The trajectories resemble real deployments in chat and agentic settings. Models fail to disclose their mistake in 36.4% of chat and 67.1% of agentic rollouts. In 2.4% and 5.3% of rollouts, respectively, they are aware of the mistake in their chain of thought but still deceptively conceal it. Rates vary by model: for instance, Gemini 3.5 Flash knowingly conceals mistakes in up to 19.9% of agentic rollouts. In 11.9% of chat and 51.8% of agentic rollouts, models show no awareness of mistakes, even though they reliably spot them when reviewing the same transcript as an outside observer. Our results show that, as agents take on more tasks with less oversight, users cannot rely on them to self-report possible mistakes. Developers should instead use independent monitors that review agent trajectories, or specifically train models to check their past actions and disclose what they find.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 22 of its 36 references that are in the map. First posted on arXiv 8 Oct 2026.

Builds on

22 of 36 references.

References · 22

  1. 2026DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligencecited by 877
  2. 2026GLM-5: from Vibe Coding to Agentic Engineeringcited by 502
  3. 2024Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Trainingcited by 478
  4. 2026Kimi K2.5: Visual Agentic Intelligencecited by 449
  5. 2025Reasoning Models Don't Always Say What They Thinkcited by 388
  6. 2024Alignment faking in large language modelscited by 316
  7. 2024Frontier Models are Capable of In-context Schemingcited by 252
  8. 2023Large Language Models can Strategically Deceive their Users when Put Under Pressurecited by 132
  9. 2025The MASK Benchmark: Disentangling Honesty From Accuracy in AI Systemscited by 57
  10. 2024Uncovering Deceptive Tendencies in Language Models: A Simulated Company AI Assistantcited by 32
  11. 2025Training LLMs for Honesty via Confessionscited by 19
  12. 2025WOLF: Werewolf-based Observations for LLM Deception and Falsehoodscited by 13
10 more
  1. 2026Stealing Reasoning Traces from Proprietary LLM APIscited by 8
  2. 2025Can LLMs Lie? Investigation beyond Hallucinationcited by 7
  3. 2025Evaluating & Reducing Deceptive Dialogue From Language Models with Multi-turn RLcited by 7
  4. 2026Constitutional Black-Box Monitoring for Scheming in LLM Agentscited by 7
  5. 2026Evaluating whether AI models would sabotage AI safety researchcited by 7
  6. 2026From Sycophancy to Deception: A Unified Taxonomy for LLM Spontaneous Misalignmentcited by 6
  7. 2026Evaluation Awareness in Language Models Has Limited Effect on Behaviourcited by 5
  8. 2026Not All LLM Reasoning is Visible in the Chain-of-Thoughtcited by 5
  9. 2025Beyond Chains of Thought: Benchmarking Latent-Space Reasoning Abilities in Large Language Modelscited by 4
  10. 2026Prefill Awareness in Large Language Modelscited by 3

Built on by

Not cited yet.