🤖 AI Summary
Rapid advancements in AI systems are outpacing the development of systematic, auditable methodologies for assessing their societal and ecological risks; current practices rely heavily on implicit assumptions and ad hoc testing. Method: We propose the first Probabilistic Risk Assessment (PRA) framework tailored to AI systems, integrating established PRA paradigms from nuclear and aerospace domains with AI-first principles. It comprises: (1) AI-specific hazard analysis; (2) bidirectional causal modeling—forward (capability deficiencies → harms) and backward (harm溯源 → capability vulnerabilities); and (3) scenario decomposition coupled with reference-scale-based uncertainty quantification. Contribution/Results: The framework enables explicit assumption tracking and absolute risk quantification. Implemented as an open-source, structured Risk Report Card tool, it delivers comparable, traceable, and quantified holistic risk estimates—supporting collaborative, high-reliability AI governance among developers, evaluators, and regulators.
📝 Abstract
Modern general-purpose artificial intelligence (AI) systems present an urgent risk management challenge, as their rapidly evolving capabilities and potential for catastrophic harm outpace our ability to reliably assess their risks. Current methods often rely on selective testing and undocumented assumptions about risk priorities, frequently failing to make a serious attempt at assessing the set of pathways through which Al systems pose direct or indirect risks to society and the biosphere. This paper introduces the probabilistic risk assessment (PRA) for AI framework, adapting established PRA techniques from high-reliability industries (e.g., nuclear power, aerospace) for the new challenges of advanced AI. The framework guides assessors in identifying potential risks, estimating likelihood and severity, and explicitly documenting evidence, underlying assumptions, and analyses at appropriate granularities. The framework's implementation tool synthesizes the results into a risk report card with aggregated risk estimates from all assessed risks. This systematic approach integrates three advances: (1) Aspect-oriented hazard analysis provides systematic hazard coverage guided by a first-principles taxonomy of AI system aspects (e.g. capabilities, domain knowledge, affordances); (2) Risk pathway modeling analyzes causal chains from system aspects to societal impacts using bidirectional analysis and incorporating prospective techniques; and (3) Uncertainty management employs scenario decomposition, reference scales, and explicit tracing protocols to structure credible projections with novelty or limited data. Additionally, the framework harmonizes diverse assessment methods by integrating evidence into comparable, quantified absolute risk estimates for critical decisions. We have implemented this as a workbook tool for AI developers, evaluators, and regulators, available on the project website.