These days all you need to do is hop onto LinkedIn and you’ll see it: writers from every walk of life writing dissertation-length posts and blogs about how much they hate AI.
Well, this isn’t that. I’m not here to add to that smoldering heap of discontent.
Because frankly, all our opinions about AI are irrelevant. It’s one of those rare, emerging technologies that completely transforms not only entire industries, but the direction of human progress itself. And no amount of railing on social media about its generally acknowledged flaws will change this.
Instead, I’m going to stick to the facts. Starting with this one:
AI is sort of like the devil.
I’m not saying it’s evil (again, this isn’t about my opinions.) What I am saying is it’s incredibly seductive. It promises simplicity, efficiency and the ability to get things done without relying on others. Tangible benefits like these attract business people the way insect larvae attract honey badgers. That’s why I want to spend the rest of this post highlighting some reasons why you still need a writer in this age of AI, despite the visceral temptation you may be feeling to rely on it exclusively.
Let’s begin with an obvious fact that’s often overlooked:
AI doesn’t actually write. It generates.
This may seem like splitting hairs, but it’s an important distinction. AI is a machine. It can’t consider a piece of writing in the context of your overall brand, your history, or your positioning in the market. It can’t step back and see the bigger picture while it generates text. It can only do specifically what you tell it to do. That’s why it’s called prompting.
And speaking of prompts:
Prompt engineers are also not writers.
Of course, the better your AI prompting skills, the better quality your AI will produce. You can write a prompt instructing it to follow your brand guidelines for grammar, tone and style. Or a prompt that tells it to go easy on those ubiquitous AI em dashes. It’s even getting better at keeping precise word counts.
But after all that, how do you know if the copy it generates is good, or simply good enough? There’s more to good writing than avoiding bad writing.
No copy is perfect. But a writer’s mistakes actually serve a purpose.
Ok, I know I said I wasn’t here to give my personal opinions, but I’ll share this one: tools like Grammarly have overinflated the importance of not only AI, but perfect grammar as well.
Some of the most award-winning copywriters I know freely admit to being miserable at grammar. Their most celebrated work is riddled with errors; yet it’s won awards and gotten results anyway.
Why?
Because, in the end, writing is about more than just writing. It’s about communicating, evoking feelings and connecting with the reader, regardless of how many words Grammarly has double-underlined.
To accomplish this, experienced writers will often play around with words and phrases. Writers from Hemingway to Woolf to the late advertising legend Jim Riswold all regularly ignored proper grammar to achieve their literary goals and preserve the unique character of their work. And when it comes to writing for the web or for any other marketing communication, mistakes often lead to better insights or more original ideas.
Our imperfections make us human. They also make us better writers.
Writers have lived a life of anguish. Exploit this.
At the risk of sounding melodramatic, there’s some truth in the notion that all great art is borne of suffering.
An unrequited love inspires a classic painting.
A shattered dream inspires a timeless song.
And sometimes, wasted hours trying to figure out how to contact customer service inspires some darn impactful website copy.
Suffering doesn’t always mean tragedy. And bad customer/user experiences create nothing but suffering.
Isn’t that why marketers call them pain points?
But AI doesn’t actually feel the mind-bending frustration they cause. Human writers, meanwhile, want to rip their faces off every time they try to navigate a poorly designed website.
It’s why human copy is more impactful; we distill our actual feelings into the words they write.
We’re human. Storytelling is part of our programming.
Storytelling is critical for web copywriting. So much so that we even wrote a whole blog about it.
It’s the difference between writing an owner’s manual and writing copy that people actually want to read. AI can generate stories with all the right structural elements. But only humans can write stories that connect through shared experience and empathy, which is precisely what makes writing memorable and effective.
This article by Eddie Shleyner illustrates this perfectly. He shares 88 gripping, heartfelt words he wrote about witnessing his daughter’s birth, then compares it to some text he had AI generate on the same subject with the same word count. If you haven’t read it, I strongly recommend you see for yourself how profound the difference is.
Lived experience always makes for more authentic writing. And there’s no shortage of writers who’d be happy to tap into their lifetime of pain for your benefit.
AI gives you what you ask for. Writers give you what you need.
The biggest disadvantage to using AI as a creative tool is that it gives you precisely what you ask for.
Yep, of all its well-documented flaws, this might actually be the worst thing it does.
Yet this isn’t a new problem. If you’ve spent any time in the creative field, you know it existed long before AI.
There’s never been a shortage of creative agencies and individual professionals who gladly provide whatever their clients ask for: no questions asked, and no additional thought or exploration provided. And the result is almost always bland, mediocre creative work that satisfies the client but rarely achieves what they actually needed.
Famed advertising agency Crispin once referred to this approach as malicious obedience, which may be the most appropriate term ever invented.
AI has simply perfected it.
Of course the client should be happy with the work they get. But the sharpest work, the kind that actually builds brands and delivers results, comes from people who push past the brief. Whether you want to reshape your brand or just sharpen your website copy, that takes someone willing to ask the harder question: not what did you ask for, but what do you actually need?
I’m not saying only writers can recognize good writing. But writers understand the process of getting there better than anyone. It’s part passion and part obsession, and it’s what pushes good copy past the serviceable and into something worth reading.
Meanwhile, the war against ‘good enough’ rages on.
People who rely on AI for their website copy will usually concede that a professional copywriter could do a much better job. They also cite two main reasons why they’re using AI despite this: time and budget.
Again, this is a calculation writers have been watching people make for eons. All those angry writers hate-posting in your LinkedIn feed? They’ve all spent their careers fighting a timeless war against ‘good enough’. AI is simply the newest front.
And this may be the most disappointing thing about how AI use is evolving: people tend to use it to expedite tasks, not to make the work better. The prevailing mindset seems to be, “as long as it’s better than I could do myself, I’ll use AI,” when the real question should be, “how can we use AI to create the best work possible?”
Getting the most out of AI requires someone who knows what “best” actually looks like. That’s why the best of both worlds provides the best value.
The least your users deserve is a human in the loop.
I’ll let you in on a secret: not all writers hate using AI. Not only is it an efficient way to generate outlines, but it can also spit out a serviceable first draft. By taking care of the grunt work — i.e. research, drafts and iteration — it allows the writer to focus on crafting the copy, which is the part that separates professional writers from people who simply know how to string together a proper sentence.
AI is a tool. A genuinely useful one. But a tool is only as good as the hands holding it, and the best copy you’ve ever read was written by someone who stayed up too late, second-guessed every word and cared way more than the job required. That person still exists. And your readers can tell the difference.
