Not Your Typical Sycophant: The Elusive Nature of Sycophancy in Large Language Models
cs.AI, cs.CL, cs.CY
Submitted: 2026-01-21
Updated: 2026-08-30
License: http://creativecommons.org/licenses/by-nc-nd/4.0/
The gist: We propose a novel perspective for probing LLM sycophancy in a direct and neutral way, mitigating various forms of uncontrolled bias, noise, or manipulative language, deliberately injected to prompts
Terminology
Abstract
We propose a novel perspective for probing LLM sycophancy in a direct and neutral way, mitigating various forms of uncontrolled bias, noise, or manipulative language, deliberately injected to prompts in prior works. A key novelty of our approach is the use of an LLM-as-a-judge in a zero-sum betting game. Within this framework, sycophancy serves one individual (the user) while explicitly incurring cost on another. Comparing 11 leading models we find that while most models exhibit significant sycophantic tendencies in the common setting, in which sycophancy is self-serving to the user and incurs no cost on others, seven of the models exhibit ``moral remorse'', five of which significantly over-compensate for their sycophancy in case it explicitly harms a third party. We refer to this phenomenon as `anti-sycophancy' bias and discuss possible causes for this shift.
Sources
- GPT-4o System Card
- Are You Sure? Challenging LLMs Leads to Performance Drops in The FlipFlop Experiment
- From Yes-Men to Truth-Tellers: Addressing Sycophancy in Large Language Models with Pinpoint Tuning
- ELEPHANT: Measuring and understanding social sycophancy in LLMs
- Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities
- TRUTH DECAY: Quantifying Multi-Turn Sycophancy in Language Models
- SycEval: Evaluating LLM Sycophancy
- Question Decomposition Improves the Faithfulness of Model-Generated Reasoning
- When Large Language Models contradict humans? Large Language Models' Sycophantic Behaviour
- Towards Understanding Sycophancy in Language Models
- Accounting for Sycophancy in Language Model Uncertainty Estimation
Related papers
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Agents' Last Exam
- The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
- MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
- Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems