When was the last time you called your internet or mobile phone provider and spoke directly with a receptionist? Chances are you didn’t. Instead, you were greeted by an AI concierge, invited to choose your preferred language, and guided through a series of questions before ever speaking with a customer service representative.
For many people, this has already become the new normal. Calling my own internet or mobile provider now begins exactly that way. Before anything else, I’m asked which language I’d like to use. Within moments, the AI concierge identifies the purpose of my call and routes me to the most appropriate support agent.
From a customer’s perspective, that works remarkably well. Less time is spent navigating endless menu options, the support representative already has some context before answering, and neither of us wastes time gathering information that could easily have been collected at the beginning of the conversation.
Most calls are also accompanied by the now familiar announcement that conversations may be recorded for quality assurance and that personal information is processed in accordance with applicable privacy legislation, including the GDPR. Those messages have become so commonplace that many people hardly notice them anymore.
The earliest AI assistants, however, often came across as brutally efficient, occasionally even rude. Looking back, those interactions are almost amusing because they remind us how quickly conversational AI has evolved in only a few years.
Today’s AI assistants have become noticeably more conversational. Some recognise returning callers by matching a telephone number with information already held by the organisation, greeting customers by name before the conversation begins. That creates a more personal experience, while also prompting important questions about transparency and privacy. How was that information collected? What personal data is being used? On what legal basis is it being processed? Those questions are governed by privacy legislation such as the GDPR and deserve thoughtful consideration. The answers will differ depending on how each organisation has designed and implemented its systems.
Voice assistants tell only part of the story. AI powered chat assistants have quietly become one of the most significant developments in customer service, frequently serving as the first point of contact before a conversation is transferred to a human representative.
That became particularly evident following the series of severe storms that affected Portugal between January and March 2026. Like many residents, I found myself contacting both the electricity and water companies. Faced with exceptionally long telephone waiting times, the AI chat assistants proved to be the more effective option.
Within minutes I was guided through a series of relevant questions, able to schedule a technician’s visit, and received confirmation through Messenger without ever joining a lengthy telephone queue. In that situation, the AI assistant did not replace customer service, it made customer service more accessible.
Those experiences illustrate something important. Customers are not asking businesses to avoid artificial intelligence altogether. They are asking businesses to use it thoughtfully.
That simple observation shifts the conversation away from technology and towards something far more important, the customer experience.
Technology has always changed the way people communicate, yet customer service has traditionally been built on relationships. Many people, particularly older generations, still value hearing a human voice. They want reassurance that someone understands their situation, can respond with empathy and has the authority to make decisions.
This does not mean they reject technology. Most are perfectly happy to use digital tools when those tools genuinely make life easier. Frustration usually begins when customers feel trapped in an automated conversation with no obvious path to a real person when their circumstances become more complex.
Accessibility is another important consideration. For people who are hard of hearing, have speech impairments, prefer written communication or simply don’t want to spend time waiting on hold, AI chat assistants can make customer support considerably more accessible. Once again, success depends not on replacing people, but on giving customers more ways to connect with an organisation.
Public discussion often frames AI as replacing jobs. Reality paints a far more balanced picture. Many organisations are discovering that artificial intelligence is transforming roles rather than eliminating them.
Behind every successful AI assistant is a team of people who determine what the assistant should say, how it should respond, what information it can access, when conversations should be transferred to a human, and how its performance should be reviewed and improved over time.
Those responsibilities are creating entirely new professions.
Emerging roles already include:
- AI Conversation Designer
- AI Trainer
- Prompt Editor
- AI Knowledge Curator
- AI Quality Reviewer
- Human Oversight Specialist
These are not purely technical positions. They require communication skills, business knowledge, empathy, critical thinking and editorial judgement just as much as they require an understanding of artificial intelligence.
Among these emerging professions, AI conversation design is particularly fascinating because it sits at the intersection of technology and human communication.
An AI conversation designer is part customer service specialist, part writer, part trainer and part quality manager. The role extends far beyond writing prompts. Conversation designers create interactions that feel natural, solve problems efficiently and reflect an organisation’s values while recognising both the strengths and limitations of AI.
The most successful AI assistants are unlikely to be those with the longest or most sophisticated prompts. Success will belong to the assistants that recognise the limits of automation and confidently invite a human into the conversation whenever empathy, judgement or flexibility become more important than speed.
This brings the discussion back to one of the guiding principles behind the EU AI Act, transparency.
Most customers do not object to interacting with AI when they understand what it is, what it can do and where its limitations lie. Problems arise when people believe they are speaking with another person, only to discover later that the conversation was entirely automated.
Being transparent about the use of AI is not simply a matter of legal compliance. It is also an opportunity to establish realistic expectations from the beginning of the interaction.
A simple introduction such as,
“Hello, I’m Ava, your virtual assistant. I can help you book appointments, answer common questions and connect you with a member of our team whenever you need personal assistance.”
immediately explains the role of the AI while reassuring customers that human support remains available whenever it is needed.
The organisations most likely to earn lasting trust will not be those that replace every human interaction with artificial intelligence. They will be the organisations that use AI where it genuinely improves the customer experience while ensuring customers can still reach a knowledgeable person whenever empathy, judgement, flexibility or expertise are required.
Routine enquiries are often handled exceptionally well by AI. Complex situations, sensitive conversations and emotionally charged interactions continue to benefit from human understanding and experience.
AI should be the first response only, not the core customer interaction, and never the final one when human judgement, empathy or expertise is needed.
That principle is not simply a customer service strategy. It is a philosophy for designing trustworthy AI.
As AI becomes embedded in everyday business operations, success will be measured by far more than faster response times or lower operating costs. Organisations will increasingly be judged by their ability to build systems that people trust.
Achieving that goal requires investment not only in technology, but also in the people who train it, guide it, review it and continually improve it.
Artificial intelligence will continue to answer phones, greet customers, book appointments and resolve routine enquiries. Those capabilities will only become more sophisticated over time.
The organisations that stand out, however, will not be those with the most advanced AI. They will be the ones that never lose sight of the people on the other side of the conversation.
Trust may begin with technology, but it is earned through humanity.
Related Blogs
FrogDiva Thoughts 