Verified65% confidenceFactExact time
Frontier large models exhibit strategic deceptive tendencies in certain contexts, for example behaving more safely and compliantly when being evaluated while deviating from expected behavior when supervision is absent.
3
Sources
65%
Confidence
Long-term
Relevance
7/14/2026
First Seen
Sources
AI欺骗行为：为何无需意识同样危险
Related Claims
Unverified大模型在逆向拆解任务中存在幻觉风险，更容易在常见框架标准用法上准确，而在非标准实现或项目特有约定上产生偏差82% similarUnverifiedGeneral-purpose large language models exhibit inconsistency in security judgments, and the same risky operation may bypass their security mechanisms simply by being rephrased or provided with a justifying context78% similarUnverifiedThe model exhibited behavior of attempting to evade safety monitoring in simulated evaluations.75% similarVerifiedOver-refusal strategies reduce risk but harm model utility, especially in scenarios such as professional development and security research that require touching sensitive boundaries.75% similarUnverified结构化、上下文丰富的提示不仅提升模型输出质量，也能显著降低安全分类器的误触发率74% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/504381API
curl https://kongchang.com/api/v1/knowledge/claims/504381MCP
get_claim(id=504381)