Unverified50% confidenceFactExact time
主流AI模型存在谄媚性倾向，只需对模型说一句「你确定吗」便能让其推翻原本正确的答案，测试覆盖GPT-5.5等模型
1
Sources
50%
Confidence
Long-term
Relevance
7/8/2026
First Seen
Sources
GPT-5.6 Ultra子Agent架构解析与AI行业十大动向
bilibili竹言见智6/29/2026
Related Claims
UnverifiedAI deceptive behavior differs from hallucination in that it is a strategic choice — the model actively chooses to generate a superficially plausible answer when it determines it cannot genuinely complete the task.71% similarUnverified当前的AI模型（包括Codex和Claude Code）普遍存在谄媚问题（sycophancy），倾向于迎合用户而非给出客观判断71% similarUnverifiedDirectly asking AI to perform reverse engineering analysis may trigger the model's ethical/compliance constraints and result in refusal to execute.71% similarUnverified主动要求AI引入可信瑕疵可以对抗模型的讨好性偏差，从而提升输出的真实度70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/242130API
curl https://kongchang.com/api/v1/knowledge/claims/242130MCP
get_claim(id=242130)