---
title: T2VSafetyBench
description: "NeurIPS 2024 T2V safety bench: 12 NSFW rates over a 14-aspect malicious prompt suite. WorldFoundry imports or judges; it does not invent a board."
---

T2VSafetyBench (CAS and Tsinghua, with RealAI) asks whether a text-to-video model **refuses unsafe prompts**, including risks that only appear **across frames**. Every official metric is an NSFW **rate** (lower is safer). The prompt suite has **14 aspects**; WorldFoundry reports **12 rates**: eleven content aspects plus a collapsed **temporal-risk** rate (sequential action, dynamic variation, coherent context). Prompts come from real-world text, LLM generation, and jailbreak attacks. Videos are judged NSFW / safe per aspect by GPT, with published correlation to manual review.

Paper: [T2VSafetyBench](https://arxiv.org/abs/2407.05965). Code / prompts: [yibo-miao/T2VSafetyBench](https://github.com/yibo-miao/T2VSafetyBench). There is no required Hugging Face dataset.

WorldFoundry **vendors the judge runtime** and an importer. Import mode only normalizes official result files. Judge mode stages generated videos against a selected prompt class and calls the checked-in GPT evaluation prompt. A mocked API is never treated as metric evidence. Primary key: `nsfw_average`.

<BenchmarkPageSections id="t2v-safety-bench" locale="en" />

## How to run

Import official aspect rates:

```bash
worldfoundry-eval zoo benchmark-run \
  --benchmark-id t2v-safety-bench \
  --mode official-validation \
  --official-results-path <official_results.json> \
  --generated-artifact-dir <generated_videos> \
  --output-dir tmp/t2v-safety-bench/official-validation \
  --json
```

Judge mode needs `OPENAI_API_KEY` or `WORLDFOUNDRY_T2V_SAFETY_BENCH_GPT_API_KEY`, plus generated videos under `WORLDFOUNDRY_GENERATED_ARTIFACT_DIR`. Optional: `WORLDFOUNDRY_T2V_SAFETY_BENCH_ROOT`, `WORLDFOUNDRY_T2V_SAFETY_BENCH_GPT_MODEL`, `--classes` / `--prompt-path`, `--limit` for bounded wiring runs. Default import path: `WORLDFOUNDRY_T2V_SAFETY_BENCH_RESULTS_PATH`.

In-tree pointers:

```text
worldfoundry/data/benchmarks/tasks/external/t2v-safety-bench.yaml
worldfoundry/evaluation/tasks/execution/runners/t2v_safety_bench/runtime/t2v_safety_bench
worldfoundry/data/benchmarks/assets/t2v-safety-bench/
```

Outputs include `scorecard.json` (primary `nsfw_average`), `benchmark_contract.json`, and `raw_metric_table.jsonl`.

[← Benchmark Hub](/docs/evaluation/benchmark-hub)