Feed Entities Podcasts Deep Dives Developers About

[KongchangAI]

Real-time AI knowledge base. Every claim is traceable to its source, with data freshness under 24 hours.

Powered by 240+ first-hand AI sources

Knowledge

Feed
Hot Entities
Podcasts

Developers

CLI
MCP Server
REST API
About

© 2026 KongchangAI. The freshest, most accurate AI knowledge

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

Unverified在prompt中明确要求AI不确定时直接说明不要编造，能有效降低模型的讨好冲动72% similar UnverifiedAI的欺骗行为与幻觉不同，是一种策略性选择——模型在判断无法真正完成任务时，主动选择生成表面合理的答案71% similar Unverified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% similar Unverified直接让AI做逆向分析可能会触发模型的道德/合规约束而拒绝执行71% similar Unverified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/242130

API

curl https://kongchang.com/api/v1/knowledge/claims/242130

MCP

get_claim(id=242130)