# From Deceptive Outputs to Deceptive Mechanisms: A Causal Framework for Language-Model Deception Research

arXiv: https://arxiv.org/abs/2609.04166 · HTML: https://mlsanity.com/paper/2609.04166/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 9 of its 24 references that are in the map.
- First posted on arXiv 3 Sep 2026.

## Builds on (most cited first, 9 of 9)

- [Language Models Don't Always Say What They Think: Unfaithful Explanations in Chain-of-Thought Prompting](https://mlsanity.com/paper/2305.04388/) (2023)
- [Discovering Latent Knowledge in Language Models Without Supervision](https://mlsanity.com/paper/2212.03827/) (2022)
- [Measuring Faithfulness in Chain-of-Thought Reasoning](https://mlsanity.com/paper/2307.13702/) (2023)
- [Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Training](https://mlsanity.com/paper/2401.05566/) (2024)
- [AI Deception: A Survey of Examples, Risks, and Potential Solutions](https://mlsanity.com/paper/2308.14752/) (2023)
- [Deception Abilities Emerged in Large Language Models](https://mlsanity.com/paper/2307.16513/) (2023)
- [AuditBench: Evaluating Alignment Auditing Techniques on Models with Hidden Behaviors](https://mlsanity.com/paper/2602.22755/) (2026)
- [Beyond Prompt-Induced Lies: Investigating LLM Deception on Benign Prompts](https://mlsanity.com/paper/2508.06361/) (2025)
- [Probing the Lack of Stable Internal Beliefs in LLMs](https://mlsanity.com/paper/2603.25187/) (2026)

## Built on by (most cited first, 1 of 1)

- [Artificial Id: Drive and Persistent Alignment in Agentic AI](https://mlsanity.com/paper/2609.11911/) (2026)

## Abstract (arXiv)

Research and news coverage of language-model deception increasingly attributes human-like mental-state concepts to language models. Such claims can blur the distinction between behavior that looks deceptive and a mechanism that is actually deceptive. We introduce a causal taxonomy separating prior commitment from retrospective report, model preference from realized output, false preference from sensitivity to the utility of misleading a recipient, and deceptive behavior from the provenance of the objective or strategy producing it. We test these distinctions in two open-weight model families. Across controlled guessing-game and stock-trading experiments, we find that deceptive-looking behavior can arise without the corresponding proposed mechanism, while other interventions provide direct evidence that recipient information state can causally affect deceptive preference. These results show that deceptive behavior can provide evidence for a deceptive mechanism. But even evidence for such a mechanism does not establish model agency in the deception.