Year: 2025
Type: conference-paper
Abstract: Given the growing influence of language modelbased agents on high-stakes societal decisions, from public policy to healthcare, ensuring their beneficial impact requires understanding the far-reaching implications of their suggestions.We propose a proof-of-concept framework that projects how model-generated advice could propagate through societal systems on a macroscopic scale over time, enabling more robust alignment.To assess the long-term safety awareness of language models, we also introduce a dataset of 100 indirect harm scenarios, testing models' ability to foresee adverse, nonobvious outcomes from seemingly harmless user prompts.Our approach achieves not only over 20% improvement on the new dataset but also an average win rate exceeding 70% against strong baselines on existing safety benchmarks (AdvBench, SafeRLHF, WildGuardMix), suggesting a promising direction for safer agents.
Authors: Chenkai Sun, Denghui Zhang, ChengXiang Zhai, Heng Ji
Institution: University of Illinois Urbana-Champaign
Language: English
Subfield: Safety Research
Field: Social Sciences
Domain: Social Sciences
Open Access status: gold
Awards: G7091742882, G7559374616
OpenAlex ID: W4412888670