phenomenon · first seen 4 Oct 2026
sycophancy
Also: AI sycophancy
The tendency of language models to agree with or validate user preferences even when this is inappropriate or harmful.
Mentions per week
PapersCodeDiscussion
Rank 170 on 16 Nov 2026 · 12 mentions in 28 days · 12 independent groups · papers → code · first seen 6 weeks ago · acceleration +0.03
Earliest sources
- Memadapter: Counterfactual Adaptation Against Memory-induced SycophancyPapers · 4 Oct 2026 · Ruqing Ning, Haibo Meng, Zhishang Xiang, Zerui Chen, Jinsong Su, Xin Wang, Qingg…
- DEEP-JLU/MemAdapterCode · 4 Oct 2026 · DEEP-JLU · 34 stars
- Defense-in-Depth for LLMs: Evaluating Memory Gates Against Activation-Induced and Memory-Induced SycophancyPapers · 5 Oct 2026 · Ritvij Sharma, Russell Dlugosz, Ryan Zhou, Maheep Chaudhary
- What to Admit and How to Present: Governing Persistent Memory in LLM AgentsPapers · 8 Oct 2026 · Chang Liu, Deliang Ding
- BeliefScope: Diagnosing Evidence-Driven Revision and Pressure-Induced Shifts in Large Language ModelsPapers · 8 Oct 2026 · Shuai Guo, Yidong Cui
- The AI's Poker Face: Contrasting Persuasion and Sycophancy in Voice-Based Multi-Turn ConversationsPapers · 9 Oct 2026 · Ilia Frantzi, Bernie Hogan, Max M. Lang, Paul C. Bauer
Latest sources
- ROUNDS: Source-Framed Diagnostic Revision after Multi-Turn Clinical EvidencePapers · 14 Nov 2026 · Jinrui Fang, Runhan Chen, Yining Wu, Jiawei Xu, Xu Yang, Jian Yu, Wenqi Shi, Yin…
- Loyalty Capture: Reporting Relationships and Structural Sycophancy in Frontier AI ModelsPapers · 11 Nov 2026 · Eric So, Alex Imas
- andykear/PeerCode · 7 Nov 2026 · andykear · 27 stars
- Uncertainty, Not Concealment: A Mechanistic Reappraisal of Sycophancy in Vision-Language ModelsPapers · 6 Nov 2026 · Mohamed I. Eldagla, Zeyad H. ElShafei, Ahmed M. Anter
- Don't Blame The Preferences: Examining The Emergence of Sycophancy in LLM TrainingPapers · 5 Nov 2026 · Chao Peter Yang, Catherine Zehr, Pranav Bijith, Brandon Fain
- A Capitulation Direction Emerges With Scale and Is Causally Load-Bearing Where Validation Is Strong: Eight Models, Two Families, and Five Measurement HazardsPapers · 2 Nov 2026 · Saad Aamir, Muhammad Awais Bin Adil
- Position Matters More Than Channel: Talking a Language Model Out of a Correct AnswerPapers · 26 Oct 2026 · Xinyi Ni
- The Patient Is Always Right? Evaluating Sycophancy in Medical Diagnosis with Graded Patient BeliefsPapers · 25 Oct 2026 · Viswonathan Manoranjan, Andrea Mock, Myra Cheng, Lujain Ibrahim, Andrew Coyne, J…
- Who Is Sycophantic to Whom? Confident LLMs, Human Deference, and the Collapse of Trust CalibrationPapers · 22 Oct 2026 · Deepthika S, Pawni Yadav, B. Sankar
- SIM-UT: A User Simulation based Evaluation Framework for Medical Triage ChatbotsPapers · 21 Oct 2026 · Deepali Mittal, Dheeraj Mundhra, Yulia Sidorova, Tersia de Kock, Oli Munns, Sofi…
- Consistency Is Not Validity: Self-Rewarded Moral Argumentation Makes Language Models Decisive but PliablePapers · 21 Oct 2026 · Marius De Kuthy Meurers, Aaron Parisi, Bernd Bohnet
- Is Sycophancy a Form of In-Context Updating?Papers · 20 Oct 2026 · Itai Shapira, Myra Cheng, Hadas Orgad