1B+ verified samples/month scraped from 19 global sources across text, image, video,
and audio, labeled by a weighted ensemble of 8 specialized AI-detection models.
anas775/DETECT-AI-Dataset/
├── text/{lang}/part-{NNNN}.parquet
├── image/{lang}/metadata/part-{NNNN}.parquet
├── image/{lang}/frames/{image_id}.jpg
├── image/{lang}/faces/{image_id}
face{NN}.jpg
├── video/{lang}/metadata/part-{NNNN}.parquet… See the full description on the dataset page:
https://huggingface.co/datasets/anas775/DETECT-AI-Dataset.
