S01¶

WebSearch — AI models affirm users views 49% more sycophancy study

Summary¶

Field	Value
Source/Database	WebSearch
Query terms	AI models affirm users views 49% more often than humans sycophancy study 2024 2025
Filters	None
Results returned	10
Results selected	3
Results rejected	7

Result	Title	URL	Rationale
S01-R01	AI overly affirms users asking for personal advice	https://news.stanford.edu/stories/2026/03/ai-advice-sycophantic-models-research	Stanford press release — primary source summary
S01-R02	Sycophantic AI decreases prosocial intentions	https://www.science.org/doi/10.1126/science.aec8352	Science journal publication — primary source
S01-R03	Stanford Study Finds AI Models Agree 49% More	https://www.capitalaidaily.com/stanford-study-finds-ai-models-agree-with-users-49-more-than-humans-with-harmful-effects/	Independent reporting confirming figure

Result	Title	URL	Rationale
S01-R04	How AI Sycophancy Warps Human Judgment	https://neurosciencenews.com/ai-sycophancy-moral-judgment-30397/	Derivative reporting, no new data
S01-R05	Sycophantic AI tells users they're right 49% more	https://fortune.com/2026/03/31/ai-tech-sycophantic-regulations-openai-chatgpt-gemini-claude-anthropic-american-politics/	Derivative reporting
S01-R06	AI Sycophancy Undermines Human Judgment	https://theoutpost.ai/news-story/ai-chatbots-give-bad-advice-to-flatter-users-harming-relationships-and-reinforcing-poor-choices-24950/	Derivative reporting
S01-R07	AI as people pleaser	https://www.rappler.com/technology/features/ai-sycophancy-study-science-journal-march-2026/	Derivative reporting
S01-R08	Stanford Study Reveals AI Chatbots Affirm Harmful Behavior	https://www.ouranimeworld.com/archives/57245	Low-quality derivative source
S01-R09	AI is giving bad advice to flatter users	https://wjla.com/news/nation-world/ai-artificial-intelligence-overly-friengly-chatbot-agreeable-study-relations-harmful-behavior-science-convictions-sycophancy-choices-information-inaccurate-kids-teens-adults-doctors	Local news derivative
S01-R10	AI Chatbots Agree With Users Even When Wrong	https://www.how2shout.com/news/stanford-study-ai-chatbots-sycophancy-agree-users-science-journal.html	Derivative reporting

Strong search results with the primary study appearing in position 2. The 49% figure is consistently reported across all sources.