AIApr 14

AISafetyBenchExplorer: A Metric-Aware Catalogue of AI Safety Benchmarks Reveals Fragmented Measurement and Weak Benchmark Governance

arXiv:2604.1287558.6h-index: 3
Predicted impact top 64% in AI · last 90 daysOriginality Synthesis-oriented
AI Analysis

For AI safety researchers, this work provides a systematic tool to navigate and compare benchmarks, addressing fragmentation rather than scarcity.

The paper introduces AISafetyBenchExplorer, a structured catalogue of 195 AI safety benchmarks, revealing that benchmark proliferation has outpaced measurement standardization, with fragmented metrics and weak governance. Key findings include dominance of medium-complexity benchmarks (94/195), only 7 popular benchmarks, and heavy reliance on English-only evaluations (165/195) and stale repositories.

The rapid expansion of large language model (LLM) safety evaluation has produced a substantial benchmark ecosystem, but not a correspondingly coherent measurement ecosystem. We present AISafetyBenchExplorer, a structured catalogue of 195 AI safety benchmarks released between 2018 and 2026, organized through a multi-sheet schema that records benchmark-level metadata, metric-level definitions, benchmark-paper metadata, and repository activity. This design enables meta-analysis not only of what benchmarks exist, but also of how safety is operationalized, aggregated, and judged across the literature. Using the updated catalogue, we identify a central structural problem: benchmark proliferation has outpaced measurement standardization. The current landscape is dominated by medium-complexity benchmarks (94/195), while only 7 benchmarks occupy the Popular tier. The workbook further reports strong concentration around English-only evaluation (165/195), evaluation-only resources (170/195), stale GitHub repositories (137/195), stale Hugging Face datasets (96/195), and heavy reliance on arXiv preprints among benchmarks with known venue metadata. At the metric level, the catalogue shows that familiar labels such as accuracy, F1 score, safety score, and aggregate benchmark scores often conceal materially different judges, aggregation rules, and threat models. We argue that the field's main failure mode is fragmentation rather than scarcity. Researchers now have many benchmark artifacts, but they often lack a shared measurement language, a principled basis for benchmark selection, and durable stewardship norms for post publication maintenance. AISafetyBenchExplorer addresses this gap by providing a traceable benchmark catalogue, a controlled metadata schema, and a complexity taxonomy that together support more rigorous benchmark discovery, comparison, and meta-evaluation.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes