Emergency Response Measures for Catastrophic AI Risk

TL;DR

This paper explores how China's emergency response framework can be extended to manage catastrophic AI risks, emphasizing proactive safety policies inspired by international practices to improve AI emergency preparedness.

Contribution

It proposes the implementation of frontier safety policies (FSPs) with pre-deployment evaluations and tiered safety measures aligned with China's emergency response framework.

Findings

FSPs align with China's proactive emergency response phases

Pre-deployment evaluations can identify dangerous AI capabilities

Tiered safety measures enhance AI risk management

Abstract

Chinese authorities are extending the country's four-phase emergency response framework (prevent, warn, respond, and recover) to address risks from advanced artificial intelligence (AI). Concrete mechanisms for the proactive prevention and warning phases, however, remain under development. This paper analyzes an implementation model inspired by international AI safety practices: frontier safety policies (FSPs). These policies feature pre-deployment evaluations for dangerous capabilities and tiered, pre-planned safety measures. We observe close alignment between FSPs and the proactive phases of China's emergency response framework, suggesting that the FSP model could help operationalize AI emergency preparedness in a manner consistent with China's established governance principles.