Utilize the Content filter of the OpenAI API to filter content (comments?)

[dkotter]

Enhancement idea.

The OpenAI API is currently in beta and as such, things might change. But they currently offer a Content filter engine that is free to use. This engine allows you to pass in text and it will classify that text as safe, sensitive or unsafe.

Thinking it would be cool to build this into Classifai as a way to determine if user generated content is safe or not. Seems most applicable for comments (think Akismet) but could also be useful on a site that accepts user content submissions, as a first line of defense.

Code of Conduct

• [X] I agree to follow this project's Code of Conduct