Using GPT API for content moderation of social app or forum

API
1

I’m a little confused as to whether we’re allowed to use the API for content moderation purposes in something like a forum or social network. I don’t mean using OpenAI’s content moderation API, I mean the GPT API.

The scenario I’m thinking of would work like this:

Prompt: Please check the following message that will be posted to a social network for any illegal, racist, offensive, explicit, or other content that isn’t PG-13 [The user’s post content goes here]

Would that be against the rules or is it ok?

2

What if the user’s post content contains things that are not allowed by OpenAI’s API terms? Should we run it through their content moderation API before using GPT as the content moderator?

3

You are not allowed to make it look like a human answers. If you make clear that an ai doesn’t like what the user posted it should be all fine.

Using content moderation upfront will help you not losing your api access.

1 Like