A prompt with multiple messages leads to better results, as the AI can learn for given examples. Alongside this change, we provide a better default proofreading prompt.
We switched to negative integers for seeded prompts, but we also
need to delete the old ones using IDs 1/2/3 if they exist,
or deploys will fail due to the unique index on the name column.
This change adds two new reviewable types: ReviewableAIPost and ReviewableAIChatMessage. They have the same actions as their existing counterparts: ReviewableFlaggedPost and ReviewableChatMessage.
We'll display the model used and their accuracy when showing these flags in the review queue and adjust the latter after staff performs an action, tracking a global accuracy per existing model in a separate table.
* FEATURE: Dedicated reviewables for AI flags
* Store and adjust model accuracy
* Display accuracy in reviewable templates
Roman Rizzi
Roman Rizzi