When a new ticket comes in, AI suggests a category, a priority, and the requester's sentiment, and flags any recent open tickets that look like duplicates or closely related requests - useful for catching a repeat report before two people start working the same issue.
From a ticket's detail page, generate a draft reply to a customer, a summary of a long back-and-forth, or a translation of any text into another supported language. Each is a starting point you review and edit before it goes out - nothing is sent on your behalf automatically.
AI can also suggest which team member is best placed to take a ticket, weighing each agent's current open-ticket load against the categories and types they've handled recently.
Every AI suggestion has a quick feedback control so you can mark it helpful or incorrect. That feedback doesn't change the suggestion on the spot, but it's a fast way to flag when the AI is off base.
Each organization has a daily and a monthly limit on AI usage across these features, keeping things sustainable at scale. Navigate to Tickets > AI Insights to see whether AI assistance is enabled for your organization and how current usage compares to both limits, broken down by feature.
Was this article helpful?