SciRisk-Bench: A Risk-Dimension-Aware Benchmark for AI4Science Safety

Researchers introduced SciRisk-Bench, a benchmark evaluating AI safety in scientific contexts across 7 disciplines and 10 risk dimensions. Tests on mainstream and science-oriented LLMs revealed where models remain unsafe, highlighting the need for risk-aware safety assessments in AI4Science.

arXiv:2606.18936v1 Announce Type: new Abstract: Large language models LLMs are increasingly embedded in AI for Science AI4Science workflows, from scientific question answering and literature analysis to laboratory planning and autonomous discovery. This progress creates an urgent need for safety benchmarks that evaluate not only scientific competence, but also whether models recognize and avoid risks in high-stakes scientific contexts. Existing AI4Science safety datasets cover several disciplines and task formats, leaving the underlying risk dimensions underspecified. We introduce \textbf{SciRisk-Bench}, a benchmark designed to evaluate AI4Science safety from two complementary perspectives: explicit risk dimensions and scientific disciplines. SciRisk-Bench covers 7 disciplines, 31 subdisciplines and 10 risk dimensions. In the experimental section, we evaluate both mainstream LLMs and science-oriented LLMs across risk dimensions, disciplines, and sub-disciplines, enabling fine-grained diagnosis of where scientific models remain unsafe.