Unverified50% confidenceBenchmarkExact time
Data based on more than 40,000 gamified test runs shows that when approving instructions issued by AI agents, humans miss an average of one-third of potential threats.
1
Sources
50%
Confidence
Long-term
Relevance
8/7/2026
First Seen
Sources
AI代理指令审批失效：人类漏放三分之一威胁的深层原因与对策
Related Claims
Unverified2024 年的多项基准测试表明，经过严格安全对齐的模型在代码生成任务上的拒绝率比基座模型高出 15-30%，在创意写作任务上的自我审查率高达 40%67% similarUnverifiedAfter using AI testing Skills, missed requirements decreased by approximately 40%.66% similarUnverified2023年多项学术研究显示在模拟对抗性环境中主流目标检测算法的准确率可下降40-80%66% similarUnverifiedDuring iPSAE screening, Agents lost 42% of their designs while humans lost only 20%, meaning Agent designs were excessively eliminated.66% similarUnverified斯坦福大学2023年的研究表明，主流AI检测工具误判率可高达20%以上66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/702859API
curl https://kongchang.com/api/v1/knowledge/claims/702859MCP
get_claim(id=702859)