Call-for-Code-for-Racial-Justice / TakeTwo

TakeTwo solution provides a quick and simple tool for media creators and media consumers to be more vigilant to detect and eliminate racial bias (both overt and subtle) in their content
Apache License 2.0
25 stars 8 forks source link

Need to define a process and crtieria for trusted user group #10

Open iainmccombe opened 3 years ago

iainmccombe commented 3 years ago

In order to prevent: (a) deliberate abuse, and (b) mins-information on what constitutes racist language,

The project requires a trusted user community to oversee the community members themselves, to critically evaluate the words and phrases submitted to the model as racist and to curate the AI model.

Critical things to define include:

  1. Criteria a user must meet in order to be considered a "trusted expert" and have the authority to review other members.
  2. An application process for users to become trusted.
  3. Outline of a process for verifying the users credentials.
github-actions[bot] commented 3 years ago

:wave: Hi! This issue has been marked stale due to inactivity. If no further activity occurs, it will automatically be closed in 14 days.

demilolu commented 3 years ago

we still need this!

naagma5 commented 3 years ago

Similar to this, we are looking into development & ML designs to promote security and robustness for trusted contributors data contributions (ie. flagging content).

demilolu commented 2 years ago

reopening - not yet complete.

github-actions[bot] commented 2 years ago

:wave: Hi! This issue has been marked stale due to inactivity. If no further activity occurs, it will automatically be closed in 14 days.