Trending
Hardly a month goes by when I fail to see something in the news about someone crossing an alleged ethical line because they used AI to produce something they wrote. In March 2026, publisher Hachette yanked a new horror novel from the US and UK markets after readers complained that it appeared to have been largely generated by AI.
Even the newspaper that first reported that incident is not immune: a reader noticed that a New York Times Modern Love column seemed to have the telltale signs of “AI slop.” The writer of the column confirmed to The Atlantic that she had used a variety of AI tools to produce the piece, though she denied simply copying and pasting prompt results into her article. Instead, she used the tools for “inspiration and guidance and correction.”
Like the rest of us, major news outlets wrestle with how to integrate AI into their writing workflows. Some readers of mainstream news organizations like The Times, The Washington Post and The Wall Street Journal revel in the opportunity to call out (usually in comments) editorials or articles that appear to have some AI content.
Ironically, a small portion—4%—of the Pope’s recent encyclical on AI may have been written by AI. While 4% may sound insignificant, it raises inconvenient questions that distract from the message.
In many, if not all, of these cases, the text was passed through the AI detector Pangram. Pangram claims an accuracy rate of 99.98%. At first, I was skeptical of this claim. The last time I looked into tools like these, they were almost as likely to produce a false positive as not.
But with time, technology gets better. In this case, a lot better.
Pangram has been rigorously tested by third parties and found to be superior to the rest of the field. That’s not to say it’s infallible. Even Pangram’s CEO says that a piece that contains a large amount of mixed AI- and human-generated text could be flagged as 100% AI content.
I’m bringing all of this up to make a point. If you are an expert who is building a reputation for thought leadership, you need to be careful how you use AI. AI can be a powerful tool, but it also has the potential to destroy your reputation.
Consider 2018 Nobel Prize winner for Literature Olga Tokarczuk. In an interview she explained how she had used AI as a research tool to help her write her latest novel. The backlash was swift. She has now decided that her current project will be her last. While Tokarczuk hasn’t admitted her chipped reputation is the reason for her decision, it is not difficult to imagine that her publisher and her readers feel deep disappointment and resentment. Much of the trust that she built up over decades has vanished in the blink of an AI.
Of course, Visible Experts® in the professional services aren’t necessarily held to the same standards as Nobel laureates. People understand that most of us are both practitioners and thought leaders. We aren’t full-time writers or speakers. That gives us, perhaps, a little more flexibility to experiment with modern technologies.
Yet with tools like Pangram readily available—a functional version is available for free—the potential for reputational damage from an agitator is higher than ever. A single post in a prominent industry forum “exposing” your “lack of original thinking” and “dependence on AI” could severely damage your credibility in an instant.
And, honestly, if you are leaning on AI to write your content, you aren’t helping your cause, anyway. As I’ve explained before, if you are letting the robots think for you, you aren’t exercising your own mental muscles. You aren’t learning in the way that writing forces you to. Nor are you producing the original, insightful stuff that readers are looking for.
In addition, because your content is machine generated, it is less likely to be considered original and authoritative by Google—making it more difficult to achieve the visibility you are seeking.
Then there’s the indelicate matter of AI-generated writing itself. It’s a mess.
As Eve Fairbanks, a professional editor, writes in a piece for The Atlantic, “AI writing is almost impossible to edit, because even when it sounds plausible, a closer look will show that every element is equally off: The tone is bland; individual word choices are baffling; the structure lacks sense; key pieces of the argument are missing; facts are false. Working on AI text, as an editor, is like trying to operate on a body whose skin, muscles, veins, bones, and organs are all compromised. There’s nothing to leave intact, nowhere to begin.”
So what is an expert’s best move? Should you completely avoid AI when you write?
I would not go that far.
I still believe AI can be an invaluable research and proofing tool. But be very careful when using it to generate text or to smooth out what you have written yourself. That’s when you enter the danger zone. If you include some AI-generated text, I suggest you revise it to sound more like you. You might even want to use a tool like Pangram to see how your piece scores before hitting the publish button.
How you use AI in writing is not a black and white issue. There is no clear line between right and wrong. Different experts are going to have different levels of AI risk. And some—those in the AI industry for instance—may actually benefit from not only using AI to write but speaking openly about how and why they do it.
But most experts are going to be far better off writing in their own, distinctive voice. If writing isn’t your jam, try pivoting to video and public speaking. Plenty of experts have built their reputations on video or in front of live audiences. Or you can simply hire a professional writer to get your ideas down on paper.
Just as the technology landscape is changing, so are the risks that come with it. I hope that now, if you’ve made it through this article, you are a little more aware of some of those risks and the decisions you need to make as you build your expert profile.
Related: Is Content Marketing Dead? The AI Search Revolution Says Not So Fast