[Error] The Prompt is unsafe: The prompt was flagged by the moderation model, stop the indexation

After applying the changes everything works as expected.

At the end of the indexation we will get an error message pointing to the logs, where we could check which content has been flagged by moderation.

See evidences:

This seems sensible to do rather than allowing proceeding. I will try to give this a test in the next several days.

Somewhat related ✨ Allow skipping of moderations for some embeddings (controlled input) Active but I think both ultimately can go in.

Thanks, gave this a good, looks good!

Automatically closed - issue fixed for 2 weeks with no activity.

kristen pol → credited marcus_johansson → .

crediting scott and marcus_johansson

A side question guys, on what basis a content gets flagged? what prompt are we talking about here? I am getting this when using Pinecone to index

@doxigo - If its openai, its the actual chunk. So if that has something that flags the OpenAI moderation.