In the age of AI, what is writing still worth?
This is a very urgent question these days. Many people who earn their living by writing are asking it.
Before we answer it, however, we must first clear up a prejudice or assumption that influences our view of the topic. Writing is not a stroke of genius. Writing is a craft and a skill that everyone can and should learn (I don’t mean the capability of putting letters together).
As with other skills, people learn it to varying degrees of proficiency. Some may only learn enough to express themselves and formulate their needs. Others develop mastery, or at least ability, over time.
Undoubtedly, a certain talent or inclination is also helpful. As with all skills, writers grow with practice and the experience gained from it.
Those who understand writing as a craft will also understand why it’s only logical that AI will keep pace with us in this skill and surpass us in many respects. Machines have learned an endless amount about writing in a relatively short time — certainly in a completely different way than we have — but one that is a thousand times superior to ours.
One human ability that AI will never achieve in this way
So, does AI render us humans superfluous when it comes to writing texts? Some fear being exposed as incapable, despite having cultivated their ability so carefully. For everyone else who previously felt they weren‘t good enough, there’s good news: AI doesn‘t take away jobs; it’s a tool that trains and improves you.
Here‘s why you won’t become superfluous as a human writer: specific experience.
No matter how much training data an AI absorbs, It remains probabilities from which algorithms generate a result. These probabilities arise from the diverse experiences people have gathered and documented.
This pool of experiences is so vast that the results produced by AI are often perceived as “creative.” The quality also depends very much on the input prompt. This is where personal experience comes into play.
Experienced writers have a multitude of data points that have shaped their work, though they might not call it that themselves:
For example, they have learned to shorten texts of various lengths in such a way that none of the actual content is lost.
If they were socialized in the print era, they most likely learned to write texts “to the line,” meaning the text fits exactly within the given framework.
They have gained experience with how their words work. They know exactly which words evoke feelings in their readers — and, ideally, which ones.
They have learned how to best structure a text so its content can be absorbed optimally.
They are familiar with all the techniques developed over decades and centuries that transform a text into something good, captivating, convincing, and informative.
Last but not least, passionate writers bring something that AIs will never possess: passion and enthusiasm for their work. Writers love their work, and that’s what makes good writing unique.
Does this mean that AI will never be able to produce such texts? By no means! Many fall victim to this misconception, either because they no longer hire writers and believe they can produce good texts with the help of AI, or because they are writers who think they have already been replaced by AI.
However: The writing profession is once again in flux. This is probably the most profound change it has experienced so far. Suddenly, writers have an AI in the passenger seat. Similar to a driving aid or autopilot, it is a tool that improves, accelerates, and elevates the overall experience to a new level.
For writers, whether they are copywriters, authors, or journalists, this means a profound transformation of their work. The core strengths of writers remain relevant:
Strategy: What do I want to achieve with my text? Who do I want to reach?
Content: What do I want to write about? What special features are worth writing (and reading) about?
Tone: How should the text sound? What feelings should it evoke?
Facts: What facts am I working with in my text?
Artificial intelligence is a good sparring partner and assistant for all these points. It can take over time-consuming activities, such as research. AI can make a text sound different in an instant — for example, it can make a text relevant to different target groups. It can also help ensure a consistent tone.
If an AI has sufficient specifications, such as a trained custom GPT in ChatGPT or a “gem” in Gemini, it can quickly formulate many different texts or transfer content into other formats. For example, it can create appealing short copy for a poster from the present text and provide five catchy headline suggestions.
No one can say exactly how AI development will progress in the coming years. Perhaps AI agents will soon be able to independently take over many more steps in content creation.
I firmly believe that, for the foreseeable future, only humans can define the core of their message — the “why.”