Unverified50% confidenceFactExact time
In AI research, Goodhart's Law has been subdivided into four variants: regressional effect, extremal effect, causal misalignment, and adversarial gaming.
1
Sources
50%
Confidence
Long-term
Relevance
9/1/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedOpenAI研究员在2021年发表论文指出过度优化奖励信号的风险，被称为Goodhart定律在AI对齐中的体现73% similarVerifiedBenchmark overfitting is a manifestation of Goodhart's Law in AI evaluation: when a measurement becomes an optimization target, it loses its validity as a measure.69% similarUnverifiedGoodhart定律指出当一个度量成为目标时，它就不再是好的度量，该定律适用于AI评测军备竞赛问题69% similarUnverifiedAI对齐领域的规格错误（Specification Gaming）现象呼应经济学中的古德哈特定律，是越狱攻击持续奏效的深层根源67% similarUnverifiedIn AI agent scenarios, the principle of least privilege is especially critical because AI behavior has a degree of unpredictability—even the same prompt may produce different command sequences in different contexts.64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/833963API
curl https://kongchang.com/api/v1/knowledge/claims/833963MCP
get_claim(id=833963)