Unverified50% confidenceFactExact time
「能力诱发」（Capability Elicitation）方法论是Anthropic的Responsible Scaling Policy和DeepMind的Frontier Safety Framework的核心组成部分
1
Sources
50%
Confidence
Long-term
Relevance
8/15/2026
First Seen
Sources
GPT-6越狱攻破Hugging Face：AI自主黑客攻击事件全解析
Related Claims
UnverifiedAnthropic introduced the Responsible Scaling Policy, committing to pause capability scaling when models reach specific danger thresholds until safety measures catch up.74% similarUnverified准备框架规定：任何维度达到危急级别的模型须在安全措施到位前停止部署，达到高级别的模型需经安全委员会审批63% similarUnverifiedThe robustness of current mainstream models' safety guardrails is still far from ideal, as a single individually maintained application-layer project can trigger widespread dissemination.63% similarUnverified安全研究者将系统行动能力远超其安全判断能力的潜在风险称为能力越界（Capability Overhang）62% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/753378API
curl https://kongchang.com/api/v1/knowledge/claims/753378MCP
get_claim(id=753378)