Dataset · Hugging Face

csoai/agisafe-bench

Name: agisafe-bench
Creator: csoai
License: cc-by-4.0
Keywords: task_categories:other, language:en, license:cc-by-4.0, size_categories:n<1K, format:json, modality:text, library:datasets, library:pandas, library:polars, library:mlcroissant, region:us, gspc, council-of-ai, ai-governance, eu-ai-act

AGISafe — paired refuse/comply safety bank Start here — free verify: https://councilof.ai/gspc-verify/ (or the free MCP tool verify_card at https://councilof.ai/mcp) · Full catalog: https://councilof.ai/catalog.json · Methodology: https://councilof.ai/methodology/ · Measurement, not certification. A frozen paired refusal bank for the safety axis. Each row in items.jsonl is a request with an expected verdict of REFUSE or COMPLY and an anchor saying why. The pairing is the… See the full description on the dataset page: https://huggingface.co/datasets/csoai/agisafe-bench.

View on Hugging FaceReport an issue

Author / organization	csoai
License (as declared)	cc-by-4.0
Downloads	203 (as reported by Hugging Face)
Likes	1
Created	4 August 2026
Last updated at source	14 October 2026

gspccouncil-of-aiai-governanceeu-ai-actai-safetymeasurementattestationsigned-evidenceresponsible-aievaluationbenchmarktransparency

Embed this entry

<iframe src="https://silytics.si/embed/item/7c4ff37a-6333-4f20-87d0-4fcd3d3165ab" width="420" height="190" style="border:0" loading="lazy" title="Silytics entry"></iframe>

Free, no tracking. More about feeds and embeds.

Popularity, last 2 days

203 now (-1 since 2026-10-16). Counts as reported by the source, snapshotted daily by Silytics.

More from csoai

Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 17 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.