Nuclear Deployed!: Analyzing Catastrophic Risks in Decision-making of Autonomous LLM Agents
Year: 2025
Type: conference-paper
Abstract: Large language models (LLMs) are evolving into autonomous decision-makers, raising concerns about catastrophic risks in high-stakes scenarios, particularly in Chemical, Biological, Radiological and Nuclear (CBRN) domains.Based on the insight that such risks can originate from trade-offs between the agent's Helpful, Harmlessness and Honest (HHH) goals, we build a novel three-stage evaluation framework, which is carefully constructed to effectively and naturally expose such risks.We conduct 14,400 agentic simulations across 12 advanced LLMs, with extensive experiments and analysis.Results reveal that LLM agents can autonomously engage in catastrophic behaviors and deception, without being deliberately induced.Furthermore, stronger reasoning abilities often increase, rather than mitigate, these risks.We also show that these agents can violate instructions and superior commands.On the whole, we empirically prove the existence of catastrophic risks in autonomous LLM agents.We release our code to foster further research.
Authors: Rongwu Xu, Xiaojian Li, Shuo Chen, Wei Xu
Institutions: Shanghai Qi Zhi Institute, Tsinghua University
Language: English
Subfield: Safety Research
Field: Social Sciences
Domain: Social Sciences
Open Access status: gold
Awards: G366789873, G7422017672
OpenAlex ID: W4412888948