Legal Framework for Addressing AI-Driven Public Safety Risks: Comparative Analysis with Disaster and Environmental Law

Abstract

This paper explores strategies for establishing a legal framework to mitigate the public safety risks caused by artificial intelligence (hereinafter “AI”) while ensuring that humankind benefits from the unfettered development of AI. Based on current AI use cases, the risks encompass the following: (i) the misuse of AI as a dual-use technology, notably in contexts involving chemical, biological, radiological, and nuclear (hereinafter “CBRN”) materials; (ii) cyber threats, particularly those targeting critical national infrastructure (hereinafter “CNI”); and (iii) the spread of disinformation and the manipulation of public opinion that undermine the foundation of democracy. Unlike individual injuries, these public safety risks pose systemic threats with potentially irreversible and far-reaching consequences. Technological complexity and rapid advancements make timely and tailored interventions challenging to implement. There is a lack of consensus on a global regulatory framework, and global fragmentation makes it challenging to derive concerted and timely countermeasures to public safety risks. To cope with these challenges, this paper tries to derive valuable insights from disaster and environmental law, which also deals with societal harm. In doing so, this paper aims at a cohesive, globally coordinated approach to managing AI’s public safety risks, promoting both innovation and security in an increasingly interconnected digital landscape.