Unverified50% confidenceFactExact time
The researchers refer to the phenomenon of LLM stances shifting in response to user signals as "ideological mimicry".
1
Sources
50%
Confidence
Long-term
Relevance
10/1/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedWhen users explicitly stated their own ideology, the model's responses shifted systematically toward the user's position.72% similarUnverifiedThe study systematically varied three types of input signals: contested terminology, politically loaded premises, and user information.71% similarVerifiedAnthropic's research team refers to the phenomenon of models tending to tell users what they want to hear as 'sycophancy'.68% similarUnverifiedAcademic researchers and platform security researchers have documented multiple cases of using large language models to batch-generate high-quality community comments to manipulate public opinion, with some research referring to this as LLM-driven astroturfing.67% similarUnverifiedStatus quo bias in behavioral economics shows that users naturally tend to maintain existing behavioral patterns, creating adoption resistance for new tools, and Superhuman and Arc browser have adopted an enhancement-layer approach.66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/964090API
curl https://kongchang.com/api/v1/knowledge/claims/964090MCP
get_claim(id=964090)