Should Rovo refuse to answer when organizational knowledge quality is too low?

[AGASTYA SHARMA]

Sometimes Rovo has enough information to generate an answer—but not enough to generate a reliable one.

Should it explicitly say:

"I don't have enough trustworthy organizational knowledge to answer."

instead of producing a potentially incorrect response?

[Dr Valeri Colon _Connect Centric_]

@AGASTYA SHARMA I’d separate prompting from a built-in safeguard. You can instruct a custom Rovo Agent to abstain when its sources don’t support an answer, but Atlassian doesn’t currently document an organization-wide knowledge-quality threshold that guarantees this behavior. A useful governance feature would be evidence-based abstention: if reliable sources are insufficient, say so, identify what’s missing, and ask for more context rather than generating a confident answer.

[Nikola Perisic]

Hi @AGASTYA SHARMA 

Yes, you should always say to the AI what to emphasize and what to avoid.