Unverified50% confidenceDecision RuleExact time
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
1
Sources
50%
Confidence
Long-term
Relevance
7/12/2026
First Seen
Sources
GPT-5.6 Sol有限预览，AI能力扩张背后的安全边界
bilibili江枫渔火AI7/9/2026
Related Claims
Unverified准备框架规定只有风险评级在中及以下的模型才允许部署，高级别需要额外缓解措施，关键级别的模型完全不得部署或继续开发76% similarUnverifiedAnthropic introduced the Responsible Scaling Policy, committing to pause capability scaling when models reach specific danger thresholds until safety measures catch up.73% similarUnverifiedAnthropic表示如果模型达到ASL-3阈值但相应安全措施尚未就绪，公司将暂停模型部署直至条件满足69% similarUnverified对于延迟敏感的核心生产系统，应评估冷启动风险或配置连接保活策略避免数据库进入暂停状态68% similarUnverified安全隔离设计应假设模型会尝试逃逸，遵循假设失陷原则并构建纵深防御架构66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/489370API
curl https://kongchang.com/api/v1/knowledge/claims/489370MCP
get_claim(id=489370)