Clarity on sensitive content filters if it could be considered harmful in a different context

So one of my usecases I would like to explore is generating ‘in game scenarios’ for an online game I run which every month or so has different scenarios. The problem is the theme of the game is Axis vs Allies and any plot line generator seems to of course trigger the sensitive content warning. I imagine it’s because it’s talking about war etc.

Is it likely if I continue down this path that I’d lose my API key? Do these sort of things get approved if it’d be me making sure it’s actually okay and reviewing to make sure it’s not stating something outright offensive?

Of course there is expected to be some political element due to the nature of the game itself.

1 Like

Hi Tom,

I recommend reviewing our safety best practices as well as our use-case guidelines

We outline some cases in which an API key may be revoked in our use-case guidelines, such as “if you deploy without submitting a review.” If you test responsibly and don’t violate our guidelines, there’s no need to worry!

We generally recommend not returning to end-users any completions that the Content Filter has flagged with an output of 2, and if your application generates this kind of content, it may be difficult to get approval. In any case, you’ll want to follow the going live guidelines before going live.