How should AI agents route exceptions to humans?

We compared two vendors on the same 20 tickets. Accuracy was fine; escalation quality was not. Score human handoff and confidence thresholds harder than model branding.

Document what “done” means for the workflow. We shipped an agent that “worked” but still required a human to close the loop every time — zero net time saved.

If you are non-technical, demand a sandbox with sample data and a 30-minute setup path. Anything that needs a solutions engineer for the first win will stall on a small team.