Short answer
It can handle part of it, the initial diagnosis. An AI agent recognizes the error, checks the state of the account and rules out known causes before the case reaches anyone. What it can’t solve is the problem nobody has seen before, which is what complex technical support is, and why a third tier exists.
In detail
Where it saves the most time
In the groundwork. On a technical case, most of the time goes into gathering information, meaning which version, which integration, what changed yesterday, what the log says. An agent that gathers all of it turns half an hour into five minutes, even when it doesn’t solve the case.
What to require from it
That it knows when to stop. A system that keeps answering a failure it doesn’t understand leaves the person with an angry customer and a conversation to undo. Here, more than with any other kind of question, it matters that it hands the case to a person with what it checked, what it ruled out and why it stopped.
The sign of a different problem
If the volume reaching engineering grows, the problem isn’t in support. Either the product is failing or the documentation doesn’t cover what people do with it, and putting an agent in front fixes neither.