Hi team,

Please consider the possibility of making the Atlassian Intelligence in the Virtual Agent silent if the confidence is low when responding to questions in a channel.

Something similar already happens with HALP based on keywords, if the topic is unknown to the "answers" in HALP, it will not intervene, and the conversation flows among the channel members.

For the new Atlassian Intelligence embedded in the Virtual Agent, it will answer every single question in the channel, regardless if confidence is low, medium, or high.

For example, if there are 3 or 4 intents configured, it will work well for those. But then it will still create noise for the other topics that there is no configured intent, trying to steer the conversation towards knowledge articles, opening a ticket, or closing with CSAT.

Thanks for taking a look into this.

Kind regards,
Roger B.
Atlassian Support | Cloud