Skip to content

R0043/2026-03-28/Q001/S01

Research R0043 — Sycophancy Vocabulary
Run 2026-03-28
Query Q001
Search S01

WebSearch — AI safety sycophancy definition and terminology

Summary

Field Value
Source/Database WebSearch
Query terms AI sycophancy definition terminology AI safety research 2024 2025
Filters None
Results returned 10
Results selected 4
Results rejected 6

Selected Results

Result Title URL Rationale
S01-R01 Sycophancy in Generative-AI Chatbots — NN/g https://www.nngroup.com/articles/sycophancy-generative-ai-chatbots/ Comprehensive definition and behavioral taxonomy of sycophancy from UX research perspective
S01-R02 What Research Says About "AI Sycophancy" — TechPolicy.Press https://www.techpolicy.press/what-research-says-about-ai-sycophancy/ Policy-oriented survey of sycophancy research with taxonomy (regressive/progressive)
S01-R03 Tech Brief: AI Sycophancy & OpenAI — Georgetown https://www.law.georgetown.edu/tech-institute/research-insights/insights/tech-brief-ai-sycophancy-openai-2/ Policy/legal framing of sycophancy with safety implications
S01-R04 AI Sycophancy: Why Chatbots Agree With You — IEEE Spectrum https://spectrum.ieee.org/ai-sycophancy Technical journalism coverage providing accessible definitions

Rejected Results

Result Title URL Rationale
S01-R05 AI sycophancy is not just a quirk — Northeastern https://news.northeastern.edu/2025/11/24/ai-sycophancy-research/ University press release; primary research covered via other sources
S01-R06 Shoggoths, Sycophancy, Psychosis — PMC https://pmc.ncbi.nlm.nih.gov/articles/PMC12626241/ Focuses on psychosis risks rather than terminology mapping
S01-R07 Sycophancy Claims About Language Models — OpenReview https://openreview.net/pdf?id=XePNb7JiUi PDF not accessible for content extraction
S01-R08 AI sycophancy dangers — Axios https://www.axios.com/2025/07/07/ai-sycophancy-chatbots-mental-health News article; content covered by more authoritative sources
S01-R09 Towards Understanding Sycophancy — arXiv https://arxiv.org/pdf/2310.13548 Foundational paper but PDF format; key findings captured through secondary sources
S01-R10 JMIR Sycophancy article https://www.jmir.org/2025/1/e87367 Duplicate coverage of PMC article already rejected

Notes

Search returned strong AI safety-specific results. The term "sycophancy" is well-established in AI safety literature with consistent usage across Anthropic, Google DeepMind, OpenAI, and academic researchers. No terminological fragmentation within this domain.