AI can’t produce the distinctive point of view that makes thought leadership worth reading—and probably won’t even save you much time. 

If you went to college in the pre-AI era, it’s likely you had that friend who would do anything to get out of actually doing their homework. Mine went so far as offering to pay me to write essays on his behalf, often the night before they were due. Which I did, once or twice, albeit to less-than-thrilling results. 

It’s obvious I wasn’t going to knock it out of the park—I didn’t know the material, and only had a few hours to get caught up. But what I remember most is sitting there with him, crouched over his desk, trying to get him to explain to me the assignment, what little he remembered from the readings he largely didn’t do, his jumbled notes. And at one point saying: “This would have been a whole lot easier if you’d just, you know, done the reading over the past few weeks, rather than trying to fake your way through it at the last second.”

In other words, I came to believe that it’s harder to BS your way through a piece of writing than to just do the underlying reading and thinking required to do it right. 

All the hullabaloo about AI-generated content—recent moves by tech giants to rid their platforms of AI slop, Anthropic’s announcement that it would watermark AI-generated text—brought me back to these experiences. So, too, has the growing belief among professional services firms that AI can do their thought leadership for them, efficiently and at the quality their clients expect. 

I’m here to say: that may not be a great idea.  

No one wants a thought leader without any thoughts  

I’m not saying AI shouldn’t be used in the writing process at all, or that I don’t appreciate why clients might want to expedite the process by teaming up with Claude or ChatGPT, then having someone like me edit it afterward. AI can be helpful in all sorts of ways when writing: in this piece, for instance, I used it to find relevant statistics (“Any recent data points support the idea that business leaders don’t like AI-generated content?”), understand what others have written about this topic, and proofread the final draft.  

The issue is that when someone offloads the actual writing of an article onto AI, it fails on two vital fronts: the quality of writing and thinking suffers and it doesn’t even save much time, if any.  

When I asked Claude to write this article for me—and I did, after I finished, to make this very point—it expressed the gist of the argument:

“The problem isn’t the tool. It’s what happens when the tool starts doing the thinking that was supposed to be the point.” 

But it did so without any of the specific examples, data points, anecdotes or insights—for instance, , using AI doesn’t even save you time!–that make this piece uniquely mine. It reads abstract, general, with all its overplayed AI-isms out in full force, resulting in a “wan little husk” of an article, to steal a phrase from Joyce Carol Oates. 

For professionals who get hired based on their expertise, that’s a pretty big problem. Greentarget’s own State of Digital and Content Marketing survey data shows that when general counsel and C-suite executives are evaluating advisory firms for hire, thought leadership is one of the biggest factors moving the needle, after trusted referrals. We also know that executives say AI-generated thought leadership content reduces their willingness to engage, and that an estimated half of U.S. consumers prefer to give their business to brands that don’t use generative AI in consumer-facing messages, advertising and content. 

Recent high-profile dustups involving hallucination-filled reports from major advisory firms only prove the point, with AI undermining the very authority the research was intended to signal.   

When and why using AI to write fails 

In my experience, there are several recurring ways that would-be thought leaders’ use of AI to write their content leads to poor results, including:  

The ‘garbage in, garbage out’ problem 

It’s true: AI can turn nothing (a query) into something (an article). But that doesn’t mean the article will be worth publishing. 

AI simply can’t create a distinctive point of view on your behalf, because it can only synthesize what already exists. It can’t tell you what’s coming next, or how to solve a pressing problem in an original way, or provide unique commentary on a new regulation or policy. Instead, you get airy, generic content that merely paraphrases what’s been said before. Quality thought leadership this does not make. 

So don’t be tempted into letting AI write something for you just because it can. Pick a topic you have expertise in and, if anything, use AI to understand what’s already been said about this and how or where you can meaningfully add to the conversation. Remember the qualities that make for distinctive thought leadership—relevance, novelty, timeliness, and utility—and try only putting something out there if you feel that only YOU could write it.  

The taste problem 

It’s hard to use AI to help you write if you don’t know what makes good writing in the first place. The result is content that’s filled with redundancies, abstraction, overly long lists, and now-classic AI tells like “It’s not this, it’s that,” sequences of three, and the inevitable one-line declarative paragraph. 

Ironically, the solution may be to look at what AI search platforms like to cite in their own answers: namely, specific case studies and/or real-world examples; compelling data that reveals the urgency of a given issue and credentials its author; and a rich, future-focused POV about a topic people actually want to hear about, now. 

The efficiency problem

When we start with a fully AI-written draft—or someone asks me to draft something out of almost thin air (with the understanding that AI can help)—are we really being more efficient? 

In most cases, I’d say no. Writing with AI not only involves time spent prompting the robot up front, but also extensive editing and iterating on the back end to make the end result sound anything close to viable as a piece of thought leadership, not to mention one without sourcing errors. 

By contrast, a solid 30-minute discovery call with a writer can generate sharper, more specific thinking and require less back-and-forth once the draft is complete. The first pass starts with something stronger and more aligned with the expert’s voice and target audience, then you tweak from there—very different than starting out with a bad piece of AI-generated content that takes substantial effort to make only a little bit better. 

All of the above is borne out by the now-familiar data points showing that 1) using AI can weaken your cognition; and 2) writing is an important mode of thinking that an overreliance on AI puts at risk. Both—cognitive debt and a hamstrung critical thinking process—are not qualities that sophisticated business leaders are looking for when evaluating a professional services firm or expert for hire.  

Let’s talk—and write—together 

At Greentarget, our goal is to help the world’s most sophisticated businesses translate expertise into impact. While we have in-depth knowledge of the subjects we write about and pitch to media, we are not the subject-matter experts themselves. 

Rather, we draw on proprietary research, a comprehensive understanding of current trends and the media landscape, and journalistic storytelling to take what is often technical or jargon-filled expertise and turn it into something that can get the attention of busy executives. 

That doesn’t happen on an AI platform. The discovery process fuels our clients’ highest quality thought leadership content: an informed conversation with an expert where we ask smart questions that help them refine, shape, and articulate a differentiated point of view—in their own words, drawing on their unique knowledge and experience. 

Put differently, we can ask the right questions. The experts we work with have the answers. And though AI can be valuable in the writing process, it can also shortcut the important dialogues and critical thinking required to create thought leadership content that can actually break through today’s morass of robot-generated noise.  

Want to talk? Reach out, here.