GPT-3 API concerned users may get me banned

Hi, I have an app that uses the gpt-3 api as a chatbot. On the open ai usage policies, it states that the api may not be used for a number of reasons such as sexual content, legal, or financial advice. My concern is that I cannot control what my users decide to use the chatbot for. Therefore - given enough users - some users will eventually try to use it for this and get me banned.

Does anyone happen to have any experience on this topic? Or have you ever been banned for this reason? What are some solutions to this problem.


Take a look at the docs when you get a chance. They’ve got some great advice for dealing with this…

Hope this helps.


The documentation that @PaulBellow provided the link for has a section about the free moderation endpoint.

You should have a read about this and use it to prefilter messages. If you are worried about latency, you might want to consider only turning it on for selected users when you see specific words in the prompts they provide. (automated would be best)