AI Incident Database - Incident 367
Source evidence cited by public records. This page shows where the source is used, its trust tier, and when it was last checked in the seed.
Records using this source
- minority-groups
entity | 4 June 2026 | 70%
minority-groups - organisation named in the AI Incident Database harm index.
- underrepresented-groups-in-training-data
entity | 14 Sept 2022 | 70%
underrepresented-groups-in-training-data - organisation named in the AI Incident Database harm index.
- iGPT, SimCLR Learned Biased Associations from Internet Training Data
event | 14 Sept 2022 | 78%
Unsupervised image generation models trained using Internet images such as iGPT and SimCLR were shown to have embedded racial, gender, and intersectional biases, resulting in stereotypical depictions.