Skip to content

R0048/2026-04-01/Q002/S01

Research R0048 — Corporate AI Training
Run 2026-04-01
Query Q002
Search S01

WebSearch — AI sycophancy in corporate/government training

Summary

Field Value
Source/Database WebSearch
Query terms AI sycophancy training corporate warning employees AI agrees with user tells what they want to hear
Filters None
Results returned 10
Results selected 4
Results rejected 6

Selected Results

Result Title URL Rationale
S01-R01 Georgetown — Reduce AI Sycophancy Risks https://www.law.georgetown.edu/tech-institute/insights/reduce-ai-sycophancy-risks/ Policy analysis of sycophancy as unaddressed risk
S01-R02 IPR — Hidden Risk of AI Sycophancy https://instituteforpr.org/the-hidden-risk-of-ai-sycophancy-in-the-workplace/ Workplace-specific sycophancy analysis
S01-R03 IEEE Spectrum — AI Sycophancy https://spectrum.ieee.org/ai-sycophancy Technical analysis of why chatbots agree
S01-R04 Science — Sycophantic AI study https://www.science.org/doi/10.1126/science.aec8352 Peer-reviewed quantification of sycophancy

Rejected Results

Result Title URL Rationale
S01-R05 Time — Problem With AI Flattering Us https://time.com/7346052/problem-ai-flattering-us/ Popular press reporting; less rigorous than selected sources
S01-R06 Northeastern — Avoid AI sycophancy https://news.northeastern.edu/2026/02/23/llm-sycophancy-ai-chatbots/ Research tips for users, not training analysis
S01-R07 Substack — Yes Man Problem https://cognitivebleed.substack.com/p/ai-research-brief-sycophancy-and Blog post; less authoritative
S01-R08 Springer — Moral Harms of Sycophancy https://link.springer.com/article/10.1007/s43681-026-01007-4 Ethics analysis, not training content
S01-R09 Stanford Sycophancy Risk https://www.techbuzz.ai/articles/stanford-study-exposes-ai-chatbot-sycophancy-risk News reporting of Science study; primary source preferred
S01-R10 Fortune — AI sycophantic bad advice https://fortune.com/2026/03/29/ai-sycophantic-bad-advice-emerging-research-science-journal/ Selected as SRC04 supplementary evidence

Notes

Search returned extensive academic and policy analysis of sycophancy but zero results showing sycophancy included in any training program. The absence is itself a significant finding.