← the record
WF-NEVL3Z

iGPT, SimCLR Learned Biased Associations from Internet Training Data

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.

Date it happened
2020-06-17
Organisation involved
OpenAI, Google
Product, system or model
Steed and Caliskan examined embeddings within image-generation algorithms, which separate pixels based on how often they co-occur within training images. Those pixel embeddings can then be used to com
Where this came from
Share this incident
XLinkedInFacebookWhatsAppEmail
Attribution

This incident was imported from AI Incident Database and is used under CC BY-SA 4.0. Our additions to it — the structured fields, the translation, the checks against other reports — are published under the same licence.

This is a record of what was reported, not a finding that anyone broke the law. If it names your organisation and you believe it is wrong, the corrections process is free and open to everyone.