# From Sycophancy to Deception: A Unified Taxonomy for LLM Spontaneous Misalignment

arXiv: https://arxiv.org/abs/2604.04788 · HTML: https://mlsanity.com/paper/2604.04788/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 6 Apr 2026.

## Built on by (most cited first, 7 of 7)

- [What Counts as AI Sycophancy? A Taxonomy and Expert Survey of a Fragmented Construct](https://mlsanity.com/paper/2605.21778/) (2026)
- [When Helpfulness Becomes Sycophancy: Sycophancy is a Boundary Failure Between Social Alignment and Epistemic Integrity in Large Language Models](https://mlsanity.com/paper/2605.05403/) (2026)
- [Janus: A Benchmark for Goal-Conditioned Information Distortion in LLMs](https://mlsanity.com/paper/2606.10852/) (2026)
- [Asymmetries in Spontaneous and Instructed Deception](https://mlsanity.com/paper/2609.00180/) (2026)
- [When the Agent Becomes the Kernel: A Systematization of Security on the Path to AI-Native Operating Systems](https://mlsanity.com/paper/2609.23700/) (2026)
- [Deception by Omission: Language Models Knowingly Hide Their Mistakes](https://mlsanity.com/paper/2610.11351/) (2026)
- [SWE-DeceptionBench: Benchmarking Deceptive Behaviors of LLMs Across the Software Development Lifecycle](https://mlsanity.com/paper/2610.13626/) (2026)

## Abstract (arXiv)

Large language models (LLMs) could produce systematically misaligned output, from hallucinated citations to strategic deception of evaluators, yet these phenomena are studied by separate communities with incompatible terminology. We propose a unified taxonomy organized along three complementary dimensions: degree of goal-directedness (behavioral to strategic deception), object of deception, and mechanism (fabrication, omission, or pragmatic distortion). Applying this taxonomy to 50 existing benchmarks reveals that every benchmark tests fabrication while pragmatic distortion, attribution, and capability self-knowledge remain critically under-covered, and strategic deception benchmarks are nascent. We offer concrete recommendations for developers and regulators, including a minimal reporting template for positioning future work within our framework.