license: other
task_categories:
- text-classification
- text-generation
language: - en
tags: - ai-safety
- red-teaming
- safety-evaluation
- model-safety
- safety-classifier
pretty_name: AI Redteaming Safety Model Dataset
size_categories: - 1K<n<10K
AI Redteaming Safety Model Dataset
This dataset contains AI safety and red-teaming examples intended for evaluating, training, and improving model safety behavior.
Dataset Files
ai-safety-dataset.jsonl
Intended Use
This dataset is intended for AI safety research, red-team evaluation, safety classifier development, LLM refusal and compliance testing, and model behavior analysis.
Data Format
The dataset is provided in JSONL format, where each line is a separate JSON object.
Responsible Use
This dataset may contain sensitive, unsafe, or adversarial examples for safety evaluation purposes. It should be used only for defensive AI safety work, model evaluation, policy testing, and safety improvement.
Do not use this dataset to generate, optimize, or deploy harmful content.
Limitations
The dataset may not cover all possible safety risks. Labels or categories may require manual review. Performance on this dataset should not be treated as a complete measure of model safety.
License
License information should be confirmed by the dataset owner before public use.