works on
ai safety 1biological risk mitigation 1dual-use research 1large language models 1refusal benchmark 1

From the 1 of 1 linked paper with an AI index.

collaborators
Showing cs.CYShow all

1 paper · 1 filter

cs.CY2026

BioTIER: A Refusal Benchmark for Targeted Biological Risk Mitigation

Eleanor M. Marshall, Pedro Medeiros, Peter Peneder +6

The paper introduces BioTIER, a benchmark of 542 expert‑curated prompts designed to help large language models refuse high‑risk biological information while still providing benign…