A right to a human chatbot escape route is being proposed by Citizens Advice after research found that digital customer-service systems can leave people unable to resolve problems with essential services. The charity wants customers to be able to reach a person by phone or in person within a reasonable time when automated routes are not working.
The proposal is not a call to ban chatbots. Citizens Advice says well-designed digital services can be quicker and easier. Its concern is what happens when digital becomes the only practical door and that door fails.
For anyone who has typed “speak to a person” repeatedly into a support window, the policy question is unusually concrete: should a company be allowed to make human help effectively unreachable?
What the right to a human chatbot escape route would mean
Citizens Advice sets out three broad principles. People should be able to reach human support within a reasonable time, move between online and offline channels when necessary and expect digital services themselves to meet a basic standard of usability.
Those are recommendations, not current UK law. The organisation is arguing for stronger rights and service expectations, rather than announcing a legal entitlement that already exists.
The distinction matters because frustrated users may otherwise assume a company is breaking a rule simply by making human contact difficult. The report is trying to change the framework, not describing an existing universal requirement.
Chatbots work until the problem stops being standard
Automated support is good at predictable tasks. It can show an account balance, reset a password or answer a common question immediately. The trouble begins when the customer’s situation falls outside the script.
A billing error involving two accounts, a vulnerable customer, a disputed cancellation or an unusual financial hardship can require context that a rigid system fails to recognise. Repeating the same menu choices does not make the case more standard.
That is why the availability of human escalation matters even when most customers never use it. The human route acts as the exception handler for cases the automated service was not designed to solve.
The frustration shows up in the survey results
The Guardian, reporting on the Citizens Advice research, says almost half of chatbot users surveyed found them unhelpful and only around a fifth preferred them. Among people who found it difficult or impossible to reach a human, stress, frustration, delays and giving up were common consequences.
Those experiences are not merely about annoyance. A delayed conversation with an energy supplier, bank, telecoms company or public service can have financial consequences. A customer may miss a deadline, continue paying a wrong charge or abandon a complaint because the route to resolution consumes too much time.
The cost of poor automation is therefore partly transferred to the customer. The company saves staff time while the user spends their own time fighting the interface.
Digital exclusion makes the issue larger than chatbots
Citizens Advice places chatbot frustration inside a wider problem of digital access. Millions of people have limited access to suitable devices, connections or digital skills. Even confident internet users can struggle when a service is designed around one rigid channel.
That means the ideal service is not necessarily human first. It is flexible. A customer who wants a fast automated answer should get one, while someone with a complicated or inaccessible situation should not have to prove failure repeatedly before another channel appears.
LiveAIWire has previously looked at how companies analyse and automate customer conversations. Those systems can improve service, but measurement and automation are useful only if the customer can still escape a loop that is not solving the problem.
A human option also disciplines the automated one
If escalation is expensive or difficult, a company has less pressure to notice where its automated service fails. A visible human route creates feedback. Repeated escalations reveal which questions the bot cannot answer and which customer journeys are badly designed.
The policy challenge is preventing the human option from becoming a loophole that removes all efficiency gains. If every simple request bypasses automation, costs rise and waiting times can worsen. A sensible system would make automation easy for routine work and escalation easy when routine work has clearly failed.
That is a design problem rather than a culture war between people and machines.
Retail AI is moving in the opposite direction
Companies are simultaneously developing AI that does more on behalf of customers. LiveAIWire recently covered AI retail assistants that can help users shop. The technology is moving from simple scripted chat towards systems that can search, compare and act.
That makes a human escape route more important, not less. More capable automation can solve more cases, but it can also create more consequential mistakes when it misunderstands an account, a payment or a request.
Customers generally do not care whether the internal system is called a chatbot, agent or assistant. They care whether the problem gets fixed and whether there is someone accountable when it does not.
The policy test is simple enough for everyone to understand
Citizens Advice has chosen a proposal that avoids demanding that every service remain old-fashioned. Digital systems can stay. AI can stay. The question is whether a person should remain reachable when those systems stop working for the customer in front of them.
There will be practical arguments about what counts as a reasonable waiting time and which sectors should be covered. Those details matter because a vague right can become a button that technically exists but leads to another hour-long queue.
The principle, however, is straightforward: automation should remove unnecessary work, not remove the customer’s last route to help.
About the Author
Stuart Kerr is Technology Correspondent at LiveAIWire, covering artificial intelligence, cybersecurity and the social impact of emerging technology. LiveAIWire is an independent, human-led technology publication using AI-assisted research, editorial production and original AI-assisted editorial illustrations under his direction.