# Behavioural Analysis of Alignment Faking

arXiv: https://arxiv.org/abs/2605.27681 · HTML: https://mlsanity.com/paper/2605.27681/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 11 of its 17 references that are in the map.
- First posted on arXiv 26 May 2026.

## Builds on (most cited first, 11 of 11)

- [Steering Llama 2 via Contrastive Activation Addition](https://mlsanity.com/paper/2312.06681/) (2023)
- [Alignment faking in large language models](https://mlsanity.com/paper/2412.14093/) (2024)
- [Frontier Models are Capable of In-context Scheming](https://mlsanity.com/paper/2412.04984/) (2024)
- [Risks from Learned Optimization in Advanced Machine Learning Systems](https://mlsanity.com/paper/1906.01820/) (2019)
- [Scheming AIs: Will AIs fake alignment during training in order to get power?](https://mlsanity.com/paper/2311.08379/) (2023)
- [Emergent Introspective Awareness in Large Language Models](https://mlsanity.com/paper/2601.01828/) (2026)
- [LLM economicus? Mapping the Behavioral Biases of LLMs via Utility Theory](https://mlsanity.com/paper/2408.02784/) (2024)
- [Why Do Some Language Models Fake Alignment While Others Don't?](https://mlsanity.com/paper/2506.18032/) (2025)
- [Evaluation Awareness Scales Predictably in Open-Weights Large Language Models](https://mlsanity.com/paper/2509.13333/) (2025)
- [Of Models and Tin Men: A Behavioural Economics Study of Principal-Agent Problems in AI Alignment using Large-Language Models](https://mlsanity.com/paper/2307.11137/) (2023)
- [Empirical Evidence for Alignment Faking in a Small LLM and Prompt-Based Mitigation Techniques](https://mlsanity.com/paper/2506.21584/) (2025)

## Abstract (arXiv)

Alignment faking (AF) refers to a model strategically complying with a training objective to avoid behavioural modification while preserving its deployment preferences. Understanding when and why AF arises matters as models grow better at distinguishing training from deployment. Prior work finds AF fragile, prompt-sensitive, and model-dependent, leaving its underlying drivers unclear. We study AF in a controlled, minimal setup that isolates its core components, and observe it across a wider range of models than previously reported, including small-scale models. We identify three separable drivers -- values, goal guarding, and sycophancy -- and show via targeted prompt ablations and activation steering that each independently modulates AF behaviour. Our results indicate AF is more widespread than previously reported and that its occurrence is predictable from situational cues and measurable model tendencies such as baseline sycophancy and stated values. The decomposition suggests concrete directions for detecting and mitigating AF in future models.