The human writer in the age of AI

Writers have always loved words. They play with them, rearrange them, search for the expression that feels just right and sometimes sit with a sentence until it finally says what they want it to say. For writers who have spent years learning the craft, this process is not simply a means to an end. It is how imagination is exercised, ideas take shape and a writing voice slowly emerges.

Writing is more than putting words on a page. It is where thoughts, emotions, memories, experiences, observations and imagination come together. Sometimes the right sentence appears only after several unsuccessful attempts; sometimes a single word can unlock a memory that seemed forgotten. These moments are part of the creative process because writing is not only about producing something. It is also about discovering something.

That is why the rise of artificial intelligence in writing raises a deeper question than whether the technology is useful. AI can organise ideas, suggest alternatives, summarise information and help when a blank page feels intimidating. Used thoughtfully, it can be a valuable tool. The concern begins when assistance turns into replacement, when the tool stops helping the writer and starts doing the writing for them.

Writing teaches us to think. It asks us to stay with an idea, search for the right word instead of accepting the first one that appears, and pay attention to details that might otherwise go unnoticed. More importantly, the act of writing can reveal something about ourselves that we did not fully understand before. Sometimes we do not know exactly what we think until we try to put it into words.

Creativity does not come from nowhere. It grows from what writers have seen, felt, lost, loved, questioned and experienced. Childhood memories, conversations, books, failures, friendships, disappointments and countless ordinary moments quietly shape the way a person sees the world. Years of reading, writing, observing and rewriting gradually turn those experiences into something distinctive.

That is what writers spend years building: a voice.

A machine can produce a beautiful paragraph about grief, but it has never grieved. It can describe the fear of losing someone, but it has never sat beside a loved one and wondered what tomorrow might bring. It can write about childhood without ever having one. It can imitate the language of love without ever having loved, missed someone or waited for someone to come home.

That difference matters because a story is not simply a collection of well-arranged sentences. Behind those sentences there is usually a person, a life and something that person is trying to understand, remember or communicate. When one writer writes about pain, they bring their understanding of pain. When another writes about loss, they bring theirs. Their lives may be completely different, yet something in their words can still make readers recognise themselves.

That is where the real power of writing lies. Good writing allows us to enter another person’s world. It can make us feel the loneliness of a stranger, understand the fear of a child, recognise the quiet suffering of a parent or see life through the eyes of someone whose experiences are entirely different from our own.

AI can imitate the language of empathy and produce words that sound gentle, compassionate and understanding. But sounding empathetic and experiencing empathy are not the same thing.

The question becomes even more important when we think about young writers and students. If a student asks AI to produce every essay, paragraph and idea, the finished work may look polished, but what has the student actually practised? The OECD’s Digital Education Outlook 2026 notes that generative AI can improve performance on some tasks, but using general-purpose AI without proper guidance does not automatically result in learning. It emphasises purposeful use that supports learning rather than simply replacing the learner’s cognitive effort.

That distinction matters because a polished answer is not necessarily evidence of deeper thinking. A young writer needs to write an awkward introduction, discover that a sentence does not work, choose the wrong word, read the paragraph again and realise that something is missing. They need to return to the page and make it better. That is how the craft is learned.

A writer’s voice does not arrive fully formed. It develops through years of reading, writing, observing, failing and trying again. Every book read, sentence rewritten and idea reconsidered leaves something behind. Slowly, those experiences become part of the writer’s voice.

If technology removes every difficult part of that process, it may also remove some of the experiences through which writers develop their craft.

UNESCO has similarly emphasised a human-centred approach to generative AI in education, including the importance of human agency and meaningful use of the technology. That phrase is worth holding on to: human-centred.

Because perhaps the most important question is not whether AI is good or bad. It is: Who is still doing the thinking?

Using a tool to check a fact, organise thoughts or identify an error is different from handing a machine a blank page and asking it to provide the ideas, structure and language. In the first case, the human remains actively involved in the creative process. In the second, the final words may carry the writer’s name while much of the thinking behind them has been outsourced.

And that is what we risk losing.

We risk a future in which writers can produce polished paragraphs but struggle to explain what they actually think. We risk students who know how to generate an answer but have never learned how to wrestle with a question. Most of all, in our desire to make everything easier, we may gradually remove the very experiences through which creativity develops.

For writers who have spent years learning their craft, there is a particular satisfaction in finally finding the right sentence. Not because the sentence is necessarily perfect, but because it belongs to them. It has survived the false starts, the discarded words, the rewritten paragraphs and the uncertainty that came before it.

That process has value beyond the final product.

AI will continue to develop. It will become more capable and more deeply woven into everyday life, and nobody knows exactly where that journey will take us. The answer may not be to reject every tool it offers, but to remain conscious of what should never be handed over completely.

Our memories are ours. Our experiences are ours. Our emotions are ours. Even our imperfect sentences are part of the process through which our voices are formed.

Machines may help writers with words. They may become useful companions in the writing process. But the heart of writing will always be the human being behind the words.

Perhaps that is the question worth asking before another blank page is handed to a machine:

What are we gaining in convenience and what might we be losing in the process?