Jul 2026
Safety, or Just Capability? A Validity Audit of Agent-Safety Benchmarks
Naming the benchmark, metric, target behavior, and model panel is the minimum a safety claim needs, and both instruments score harmful compliance, so this is evidence of convergent validity rather than general safety.
You Wang, Xiao Han, Ding-Yan Shang et al.
· arXiv.org · 1 citation