There is a peculiar pattern in the working lives of those of us who have been around long enough to remember when office technology meant a typewriter, a calculator and a filing cabinet. Every few years, the tools change, the terminology changes, and suddenly we are expected to prove all over again that we are capable of doing our jobs. First, we had to prove that we could type. Then we had to demonstrate that we could use word-processing software. Next came spreadsheets, databases and customer relationship management systems. Now, the latest requirement is AI fluency, often accompanied by talk of prompt engineering, AI-powered workflows and the need to keep up with the next generation of tools. The technology has changed enormously, but the question is whether our underlying professional abilities have changed as dramatically as the demands placed upon us suggest.
The Moving Goalposts of Workplace Competence
Think back to the 1980s, when typing speed and accuracy could be important requirements for an office position. Being able to type was a practical skill, and employers had good reason to want people who could do it efficiently. As word-processing software became commonplace, however, the ability to operate a typewriter was no longer sufficient. We had to learn new systems, new commands and new ways of producing documents. Then came the spreadsheet revolution. Accounting, budgeting and financial planning increasingly moved into Excel and similar applications. Suddenly, being comfortable with numbers was not necessarily enough; employers wanted to know whether we could manipulate data, use formulas and produce reports. As organisations became more reliant on databases and customer relationship management systems, another requirement emerged. Could we navigate the CRM, maintain records, generate reports and manage customer interactions within the prescribed system?
The timeline varies by industry, country and job role, but the pattern is familiar: typing, word processing, spreadsheets, databases, CRMs and now AI tools. Each technological shift brings genuine opportunities, but it also seems to bring another test of employability. Of course, learning new tools is part of professional life. I would not want to return to the days when producing a document meant correcting a typewritten page or manually reworking calculations that a spreadsheet can now perform in seconds. Technology has made many tasks faster, easier and more accessible. My concern is not that employers expect us to adapt. It is that familiarity with a particular tool can become a proxy for competence, while the skills that make someone good at a job receive less attention. Knowing how to use Excel does not automatically make someone a good accountant, just as knowing how to navigate a CRM does not make someone an effective customer service professional. And knowing how to write a prompt does not automatically make someone a good writer, researcher, analyst or strategist. The tool and the skill are not the same thing.
First ChatGPT, Then Claude. What Comes Next?
When ChatGPT first entered the public consciousness, there was an almost frantic rush to learn how to use it. Courses appeared, certifications were offered, tutorials proliferated, and suddenly everyone seemed to be an aspiring prompt engineer. People explored the technology, experimented with its possibilities and tried to understand how it might change their work. Much of that enthusiasm was justified. AI opened up new ways to generate ideas, summarise information, draft content and tackle tasks that previously required considerably more time. Several years later, we sometimes find ourselves asking a question that would have seemed absurd before the technology became commonplace: how did we manage before ChatGPT?
The answer, of course, is that we had Google. And before Google, we had encyclopaedias, libraries, card catalogues, reference books and, occasionally, a knowledgeable librarian who knew exactly which shelf to point us towards. Imagine that. We searched, read, compared sources, took notes and assembled our findings. Sometimes it took longer. Sometimes we had to consult several books or chase a reference through a maze of footnotes. But the underlying task was the same: to find information, assess what we had found and make sense of it. The tools changed, but the need to think critically did not.
Now the attention is shifting again. Claude and other AI systems are attracting their own followers, courses and claims of indispensability. Tools that were considered essential only yesterday are suddenly described as outdated, while the next platform promises to transform the way we work. Even Canva, once celebrated for making design accessible to almost everyone, finds itself operating in an environment where new tools and AI-powered alternatives appear at remarkable speed. Meanwhile, some of us are still making PowerPoint presentations. And apparently, that is enough to provoke the question I recently heard in a meeting: “Who even uses PowerPoint now?” Ouch.
The irony is that PowerPoint remains useful for presentations, just as Google remains useful for searching and Canva remains useful for design. A newer tool may offer different capabilities, faster workflows or more convenient automation, but newer does not automatically mean better for every task or every user. There is a difference between learning a new tool because it genuinely improves our work and learning it because we fear being considered obsolete if we do not. When every new platform becomes another test of employability, we risk spending more time proving that we can use tools than demonstrating what we can actually accomplish with them. Technology is supposed to help us work. At some point, it is worth asking whether we are still choosing tools to serve our work or choosing our work around the tools we are expected to know.
We Learned to Think Before There Was an App for Everything
My own experience in development work during the 1990s offers a useful example. In those days, we ran workshops using a German facilitation method known as Metaplan. The kit came with cardboard cards and shapes on which participants could write their ideas. We could arrange the cards into groups, map relationships, identify problems and priorities, and gradually develop the material into coherent plans, statements or mission statements. I loved it. When I ran workshops with Tibetan groups in the Himalayas, the method was particularly effective because it was tactile and straightforward. Participants could see the ideas, move them around, discuss them, challenge them and build a shared understanding. They could also learn the process and replicate it themselves without needing sophisticated equipment or specialised software.
The point was not the cardboard. It was what we did with it. We were facilitating discussion, organising information, recognising connections and helping people turn a collection of individual contributions into a shared outcome. We were planning projects and solving problems. The method gave us a practical way to make thinking visible. Today, we have digital whiteboards such as Miro, FigJam and Mural, where virtual sticky notes can be moved, grouped and connected across continents. They are useful tools, particularly when teams cannot gather in the same room, and some can even use AI to help organise or summarise contributions. Yet the essential work remains remarkably familiar. Someone still has to ask the right questions, recognise when ideas belong together, identify contradictions, encourage participation and help the group decide what matters. We have digitised the cards, but we have not digitised the need to think.
The cardboard cards were not inherently superior to their digital equivalents. The advantage was that the method matched the people, the task and the context. It made a complicated process accessible without making the participants dependent on the technology. That distinction matters. A competent professional should be able to learn a new platform when the work requires it, but the ability to facilitate a discussion, structure an argument or understand a problem does not suddenly appear when someone opens a digital whiteboard. Nor should it disappear from an employer’s assessment of competence simply because the work is now carried out through a different interface.
We Have Never Been More Connected. So Why Does Work Sometimes Feel Less Human?
The changes in our tools have affected more than how we complete tasks. They have also changed how we interact with one another. Before the pandemic, many of us worked in offices. If I had a question, I could walk down the corridor and ask a colleague. We might have a quick conversation, a longer discussion or even an argument. We could sort things out there and then, read each other’s reactions and move on. Sometimes the conversation wandered off topic and produced an idea nobody had been looking for. That was part of the process. We also learned from the informal exchanges that rarely appear in project plans. We overheard conversations, picked up useful information from other departments, noticed when a colleague was struggling or discovered that someone nearby had already solved the problem we were trying to tackle. Much of this knowledge moved through an organisation without anyone formally scheduling it.
Over time, conversations increasingly moved to email, corporate intranets, Slack and Microsoft Teams. Meetings moved online, and remote working became an established part of professional life. I have worked on two projects where I never met my team members in person throughout the entire time I worked with them. We communicated, coordinated our work and got things done, yet we did all of that without ever sharing a room. There are obvious advantages to this evolution. Teams can work across countries and time zones, information can be documented and shared, and people no longer have to be physically present in an office to contribute. For many people, remote working has opened opportunities that would otherwise have been inaccessible. But something changes when so much of our interaction is mediated by technology.
A written message carries words, but not necessarily tone, facial expression or the small signals that help us understand one another. A virtual meeting can bring people together, but when cameras are switched off, participants may be little more than names on a screen. And the informal exchanges that once happened naturally in an office now require someone to initiate a message or schedule a call. A question that once required a ten-second walk may turn into a series of messages, a delay while someone responds, a misunderstanding and eventually a meeting to clarify what could perhaps have been settled in a brief conversation. We have become very good at transmitting information. Whether we have become equally good at creating human connection is a different question.
This is not an argument for returning to an idealised past. Offices could be noisy, distracting and full of unnecessary interruptions. Remote teams can build trust, collaborate effectively and produce excellent work. Nor does every conversation need to happen face to face. The issue is whether we recognise that communication technology is not a substitute for the human skills that make communication work. Work involves more than completing assigned tasks. It involves trust, negotiation, mentoring, learning from colleagues and navigating disagreements. Some of those things happen in formal meetings, but many develop through ordinary, unplanned interaction. You cannot put every useful conversation into a calendar invitation, and not every misunderstanding can be resolved by adding another message to an already crowded channel. Being connected is not quite the same as feeling connected.
If we want technology to make our working lives better, we need to consider not only what it enables us to do, but also what we need to preserve in the way we work together. This becomes particularly apparent when we look at customer service, where efficiency is increasingly measured by how quickly a system can process a request rather than by the quality of the interaction itself.
Hello, AI Assistant Max. May I Speak to a Human Being?
AI assistant Max may be efficient, endlessly patient and available around the clock. In some countries, voice systems are becoming so sophisticated that callers may struggle to tell whether they are speaking to a human being or an AI. But I still prefer a Frau Meyer or a Herr Schmid on the other end of the telephone, someone who can answer my question and, in the course of the conversation, perhaps discuss the football results or complain about the weather today. That little bit of unscripted conversation may not appear on any efficiency chart. It may not shorten the call or improve a customer service metric, but it reminds us that communication is more than the exchange of information. It is also a social interaction, an opportunity to connect with another person, even briefly.
Of course, nobody wants to spend twenty minutes listening to hold music or explaining the same problem to five different departments. Automation can make customer service faster and more accessible, and a well-designed AI assistant can be genuinely useful. The problem arises when efficiency becomes the only objective and human contact is treated as an unnecessary expense. Sometimes I want the answer to my question. Sometimes I also want to speak to another human being. Those are not mutually exclusive requirements. Perhaps the real measure of progress is not whether an AI can imitate a human conversation so convincingly that we cannot tell the difference, but whether we have made our services more efficient without forgetting why human interaction still matters.
There is another dimension to consider. Some voice systems invite callers to provide a voice sample so they can recognise them during future calls. The convenience is understandable, but it also raises questions about what information we are handing over, how it is stored, how long it is retained and how it might be used in the future. Depending on how a voice sample is processed, it may qualify as biometric data used to identify a person, which receives special protection under the EU’s GDPR. Not every voice recording automatically qualifies, and the specific obligations depend on the circumstances. Still, it is reasonable to ask questions before handing over information that could become part of a system designed to recognise us. Learning to use technology should also mean understanding what we are agreeing to when we use it.
AI Fluency Is Not the Same as Professional Competence
AI makes the distinction between operating a tool and exercising professional judgement particularly important. Knowing how to operate an AI tool is useful. Understanding what it can and cannot do is even more useful. But neither guarantees that someone can assess the quality of its output, recognise missing information, identify an invented claim or decide whether a particular task should be delegated to AI in the first place. AI fluency should involve more than writing effective prompts. It should include critical evaluation, an understanding of limitations, appropriate handling of information and the judgement to know when human expertise must take precedence.
A person who understands a subject deeply may be better equipped to use AI responsibly than someone who has memorised a collection of prompting techniques but lacks the knowledge to evaluate the answers. The same principle applies to every technological shift that came before. Software can make a task easier, but it cannot automatically supply the experience, judgement or accountability required to do that task well. Of course, some jobs genuinely require specialist knowledge of particular platforms. An employer is perfectly entitled to ask whether a candidate can use the tools necessary for the role. But there is a difference between assessing a relevant skill and treating every new tool as a universal measure of professional worth.
Employers should be able to explain what they mean by AI fluency, why it matters for a particular position and how they will recognise it. Is the requirement to generate content, automate a workflow, analyse data, evaluate AI output or understand the risks involved? Those are different capabilities, and they deserve different assessments. A certificate may demonstrate that someone completed a course. It does not, by itself, demonstrate that they can apply what they learned intelligently in a real working environment. Experience matters. So does curiosity, adaptability and the willingness to learn. But these qualities should be recognised as part of a person’s professional competence, not continually ignored until they acquire the next fashionable credential.
Perhaps It Is Time to Change the Question
I have worked through several generations of workplace technology. I have learned new systems, adapted to changing expectations and continued to explore tools that might make my work more effective. I am not opposed to technological progress, and I have no desire to abandon the advantages it offers. But I did not begin learning how to think when the first digital tool arrived. I learned to plan before planning had an app. I learned to organise information before AI could summarise it. I learned to facilitate discussions before virtual whiteboards existed. And I learned that getting a group of people to work towards a common goal requires more than putting their ideas on a screen.
The technology will continue to change. There will be new AI systems, new collaboration platforms, new certifications and new claims that the latest tool is indispensable. Some will bring meaningful improvements. Others will prove useful only for particular tasks. We will have to make our own judgements about which ones deserve our time. Adaptability is not the same as chasing every trend, and being willing to learn is not the same as having to prove one’s worth from scratch every few years.
Perhaps employers need to look beyond the list of platforms a candidate has used and ask better questions. Can this person analyse a problem? Can they communicate clearly, work with others, learn from experience and make sound decisions? Can they recognise when a tool is useful and when it is getting in the way? Can they adapt when circumstances change? These are not obsolete skills waiting to be replaced by AI. They are the capabilities that allow us to use technology effectively in the first place.
The challenge is not to stop learning new tools. It is to stop confusing the tools with the people who use them. We can replace the typewriter with a word processor, the spreadsheet with a more powerful analytical platform, the cardboard cards with a digital whiteboard and the office meeting with a video call. We can ask AI to help us draft, summarise, organise and analyse. But we still need people who understand what they are trying to achieve, why it matters and how to recognise a good result.
So, before we ask candidates whether they are fluent in the latest platform, perhaps we should ask what they have learned, what they can contribute and how they approach a problem they have never encountered before. And perhaps those of us who have spent decades adapting to new technologies should stop being asked to prove, yet again, that we can learn how to use a tool. How many times must we prove we can use a tool before employers recognise that we know how to think?
Related blogs:
It’s AI! What Does That Actually Mean?
