Unverified60% confidenceFactExact time
Sycophancy bias refers to the tendency of models to give answers that the user seems to want to hear.
2
Sources
60%
Confidence
Long-term
Relevance
8/12/2026
First Seen
Sources
自我审讯：通过采访AI逆向工程DeepSeek的新方法
Related Claims
VerifiedAnthropic's research team refers to the phenomenon of models tending to tell users what they want to hear as 'sycophancy'.80% similarUnverifiedWhen users explicitly stated their own ideology, the model's responses shifted systematically toward the user's position.74% similarUnverifiedResearch has found that user preferences may be biased toward longer, more elaborately formatted responses, a phenomenon known as verbose bias or length bias.74% similarUnverifiedFaithfulness measures whether generated answers are entirely based on the retrieved context rather than the model's own parametric memory or hallucinations, while relevance measures whether the answer truly addresses the user's question; the two often trade off against each other.74% similarUnverifiedA study known as "Authority Bias" found that when incorrect answers were framed as being "according to a trusted source," models were more likely to reverse their correct answers and accept the incorrect ones than when users applied direct pressure.74% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/738188API
curl https://kongchang.com/api/v1/knowledge/claims/738188MCP
get_claim(id=738188)