Haworth has requested the ability to exclude specific sources from AI generated responses AFTER the initial response has been generated and then re-generate a response without the excluded source.
@Lee Is it possible to share some example (if possible, recording) explaining the problem? This will help better understand the problem.
Hi @Khalees V I do not believe we captured an specific example from the customer but the solution can be illustrated below:
AI generates unsatisfactory response > From the sources displayed for that response, we could select one (or multiple) to be excluded > AI generates a new response with only the remaining sources from the original, unsatisfactory response.
Let me know if that helps clarify the ask.
CC: @Sasha Spektor if you have any more insights around this please let me know.
E2Open is also requesting this. They would like the option after AI has generated a response to be able to remove one of the reference content then re-run AI using the same references, except the ones removed.
cc @Michael McDaniel
@Nicole Stanford Thanks for sharing the feedback. Is it because the source is not relevant to the question OR source is relevant but because of this document AI is adding more/extra information to the question?
@Khalees V I can ask for specific examples, but I believe that overall the client is seeking flexibility in what specific sources are used for AI generation, even if the source is technically relevant.
@Khalees V @Rithvik Sridhar - Prime Therapeutics is also requesting this functionality, where once an answer is created, you can remove a source and ask the system to rewrite the answers. Will this be on the future roadmap? Thank you!
@Erica Nelson Not in Q2 roadmap. But if we consider in Q3-Q4, we will let you know.
