BioTIER Evaluates Biosafety Guardrails
Miles_Brundage · x · 2026-07-17
SecureBio has introduced BioTIER (Biological Targeted Information for Exclusion and Refusal), the first evaluation benchmark designed specifically for biosafety-related guardrails.
This benchmark aims to fill a gap in the evaluation ecosystem by measuring both safety and usability, rather than solely focusing on whether a model refuses to answer. The authors highlight three key findings:
- Refusal behaviors vary significantly across different models
- The strongest guardrails are concentrated in a few closed-source models
- Model scores on BioTIER fluctuate over time
Full details are available via the provided link.
More from Safety
- Ex-DOGE cyber researchers breach Anthropic systems with AI agents in under 12 hours — ScottNover · 2026-10-26
- Ofcom admits UK online safety laws failed to protect children in first review — gnukeith · 2026-10-26
- Sakana AI CEO Calls for Japan-US Shared Framework for AI Capability Evaluation — SakanaAILabs · 2026-10-26
- Gary Marcus backs call to outlaw voice cloning, citing scam and fraud harms over benefits — GaryMarcus · 2026-10-26
- Turing-complete malware can run in PyTorch's backward pass via a zero-valued loss — retr0jirachi · 2026-10-26
- Import AI 478: $135k math autoformalization, ProgramBench hits 18.5%, agents leak via CPU cache — Import AI (Jack Clark) · 2026-10-26