AT&T Customer Says AI Chatbot Turned a Simple Repair Request Into a Frustrating Loop
AI customer support is supposed to make life easier. For major U.S. carriers such as AT&T, chatbots are often promoted as a faster way to answer questions, troubleshoot service issues, and reduce wait times for subscribers. But for one customer, a basic request for help reportedly turned into a frustrating back-and-forth with AT&T’s Andi chatbot that went nowhere.
The issue itself was simple: the customer’s dog had chewed through a line connected to the router, and he wanted to know whether AT&T could send a technician to repair or replace the damaged cable. In theory, this should have been a straightforward service request. Instead, according to the customer’s shared experience on Reddit, the chatbot repeatedly failed to move the conversation toward a real solution.
The customer tried explaining the problem and asked to speak with a human representative. However, Andi appeared to keep circling through automated prompts rather than properly recognizing the request. What should have been a quick appointment setup became a long and exhausting exchange, with the customer stuck trying to convince the system that he needed actual support.
This kind of experience highlights a growing frustration among consumers as companies increasingly rely on AI-powered customer service tools. While chatbots can be useful for simple questions, they can become a major obstacle when a customer’s issue falls outside a narrow set of expected responses. In those situations, users often just want a clear path to a human agent.
The incident also sparked debate among commenters. Some users criticized the customer, suggesting he should have used shorter messages or gone through AT&T’s Smart Home Manager app instead of explaining the problem in detail to the chatbot. Others argued that the burden should not be on customers to “learn” how to talk to an AI system just to report a damaged internet line.
That reaction points to a bigger problem with automated support. If a chatbot requires users to phrase requests in a very specific way before it can help, then it is not really simplifying the customer service process. A good support system should understand common real-world problems, especially something as basic as a damaged cable needing technician repair.
AT&T is not the only carrier facing criticism over chatbot support. Customers across the telecom industry have complained about automated assistants that provide generic answers, repeat irrelevant suggestions, or make it difficult to reach a live person. Companies may see AI as a way to improve efficiency, but customers often judge these systems by one simple standard: did it solve the problem?
In this case, the answer appears to be no.
AI tools can still have a valuable role in customer support, especially for billing questions, plan comparisons, outage checks, and simple troubleshooting steps. But when a subscriber needs a technician, repair appointment, or account-specific help, the system should quickly identify that need and escalate the conversation. Otherwise, the chatbot becomes less of a support tool and more of a barrier.
For AT&T customers dealing with damaged equipment or home internet issues, the safest approach may be to use the company’s official app, check for repair options in the account dashboard, or try reaching phone support if the chatbot fails to understand the request. Still, customers should not have to jump through endless hoops to get help with a basic service problem.
The situation is another reminder that AI customer service still has a long way to go. Faster responses mean very little if the answers are not useful. For subscribers, the best support experience remains the one that actually fixes the issue, whether it comes from a chatbot or a human being.
