All those lists of AI identifiers doing the rounds these days may seem accurate to an extent, but to rely on them to give you a thorough rundown on the state of AI-generated content would indeed be a wrong move.
If you doubt this, all you have to do is pick up one of these popular words or phrases and see how many times they’ve featured in human-written content that’s been setting the bar even before the advent of AI. A couple of exercises down this lane is all you’ll need to know this to be true.
So how do we identify content written by AI tools? Or, more importantly, how do we ensure we do not end up tagging purely human-written content for the wrong reasons, only because they happen to feature on some list?
Well, we asked our content experts, business leaders, and marketing community for answers, and although this list (or any other) may never be enough to draw a line, it pretty much narrows the gap!
Read on!
AI Repeats Phrases Unnecessarily
One of the most amusing identifiers for AI-generated content is the claim that “it repeats phrases unnecessarily.” While it’s true that some AI models lean on redundancy, humans are just as guilty, especially in casual writing or overly enthusiastic sales copy. I’ve seen emails from actual people where “innovative” appeared five times in two paragraphs, and no one flagged that as AI!
On the problematic side, the criticism that “AI lacks personality” feels outdated. AI-generated content can now mimic tone, humor, and even quirks when trained on the right inputs. The real challenge lies in context and nuance. An AI might write, “This solution guarantees success!” – a phrase that lacks the caution or hedging professionals often include, which could mislead readers.
These identifiers are helpful for analysis but often oversimplify the complexities of writing. AI-generated content isn’t inherently bad- it just needs human editing to add that final layer of finesse.
Inge Von Aulock
Founder & COO, Penfriend
Overuse of 'By' and 'Delve'
I’ve noticed that phrases like “By…” and words like “Delve” or “Dive” tend to be overused by AI-generated content, and they’re often treated as clear indicators. In my opinion, phrases like “By doing X, you’ll achieve Y” have become so formulaic in AI outputs that they immediately stand out.
In the same way, “Delve” and “Dive” can sound elegant but feel forced when used over or excessively. These terms aren’t inherently bad, but their repetitive use makes them feel generic, which can be a dead giveaway. I believe they speak to how AI tends to settle for logical, secure solutions with general utility, rather than what humans would naturally choose to say.
Honestly, I don’t see anything wrong with using identifiers to check for AI content; if we want AI-written material, we’d just generate it ourselves. Paraphrasers and human intervention can refine content further, ensuring that it feels more authentic and tailored to the audience. For me, the real skill lies in blending AI efficiency with human creativity.
Matthew Goulart
Founder, Ignite Digital
Misunderstanding AI Capabilities
I find the obsession with “identifying” AI-generated content simultaneously amusing and problematic. Terms like “robotic tone” or “too perfect” always get an eye roll from me because they highlight the misunderstanding of what AI can and can’t do. AI doesn’t always sound robotic anymore. In fact, some of the most polished, conversational content is AI-generated, and the irony is that striving for “too human” is precisely what gives it away sometimes.
The idea that using AI negates your creativity or insight bothers me. It’s not the tool itself that’s the problem, but how it’s used. A tool like ChatGPT can generate bland, formulaic output if prompted poorly. I can’t imagine any quality writer using AI exclusively and without editing. It is a tool, not a complete replacement for a human writer.
In my line of work, where storytelling and messaging are critical, I see these misconceptions as an unnecessary fear. The amusing part is that people will spend so much time trying to “identify” AI-written content when they could use the same technology to enhance their creativity and productivity. It’s not about the content being “too perfect”-it’s about whether it is correct, resonates with the audience, and serves its purpose. That’s what matters most.
Christine Wetzler
President & Founder, Pietryla PR & Marketing
False Positives in AI Identification
In my opinion, relying on specific terms or phrases to identify AI-generated content is not the right approach.
This is because LLMs are trained over massive datasets with different styles and tones. Therefore, if any of the “identifiers” match the training data or overlap the human-generated content, then it can be falsely claimed as AI-generated content.
The bigger problem, though, is dismissing the value of content by saying it’s “AI-generated.” It can seriously limit your potential.
As for amusing identifiers, one of our clients claimed that our writers used AI to write blogs because we used words like “facilitate, jeopardize, reluctant, and contemporary.”
Gursharan Singh
Co-Founder, WebSpero Solutions
AI Writing Too Consistent
The most amusing “identifier” is probably the claim that AI writing is too consistent or formal in tone. As we can see from all our varied writing examples, AI can adapt tone significantly – from professional to casual, technical to conversational. The idea that AI only writes in one voice misses how language models actually work.
I also find the focus on phrases like “In conclusion” or “Let me explain” problematic as AI identifiers. These are common transitions used by human writers too. Attributing them solely to AI suggests confirmation bias – people see what they expect to see.
Similarly, claims about AI always using the most common word choices or having perfectly consistent punctuation don’t hold up to scrutiny. Language models can employ varied vocabulary and natural variation in formatting, just as human writers do.
The underlying issue is trying to identify AI content through superficial patterns rather than examining substance and quality. What matters more is whether the content is accurate, helpful and ethically created – not whether it contains certain phrases or follows rigid patterns.
Lisa Benson
Marketing Strategist, DeBella DeBall Designs
Buzzwords and Formal Language
I’m amused by the idea that repeating certain buzzwords or using overly formal language signals AI. Personally, I find it problematic because some human writers sound official too. It’s not a solid indicator, just a stereotype that assumes all bots write stiffly and humans always stay casual.
Modestas Mankus
Founder, Talks Grow
Overuse of Formal Language
Common identifiers for AI-generated content, like ‘overuse of formal language’ or ‘lack of logical flow,’ can be both amusing and problematic. Amusing because AI has grown sophisticated enough to mimic human quirks and creativity, making these traits unreliable markers. Problematic because this mindset underestimates the potential of AI as a creative collaborator. Instead of focusing on ‘identifying’ AI content, the emphasis should be on the quality, accuracy, and relevance of the message-regardless of who (or what) creates it.
Samanyu Marda
Marketing Intern, Mailmodo
Common AI Repeated Annoying Phrases
Common AI repeated annoying phrases:
- You’re going to find out
- I’m going to walk you through
- Don’t worry too much about
- You can always adjust your approach down the road
- Your first attempt doesn’t need to be your last
Of course, words such as delve, embarking, pivotal, quest, unravel, and game-changer, just to name a few, are very problematic too.
However, the more updated LLM models are getting better at removing these annoying, lame words and phrases. But, as we all know, AI has a habit of hallucinating so you need to constantly check your work.
In the beginning, I hardly noticed this strange phenomenon of misguided words being thrown onto the screen. Now, I can spot them in a New York minute! Proofreading my own work and the work of others has been imperative.
A much better way is to use talk to text where at least the words being written out are your own. Then you can go back and edit accordingly. In this way, you can bypass any computer generated AI jargon.
Structural Problems in AI Content
Apart from the all too well-known ‘let’s delve into the topic of..’ it is of course also a structural problem. Ask ChatGPT to write you an article about anything and it is guaranteed to end in something like an H2 heading ‘Conclusion – ‘ or a very generic similar text-structural feature. The very best articles/content I read very rarely have this, it is not a must-have for good content, but an overly formalistic approach.
So I personally identify AI content often by the monotony and at the same time precise structure of it. AI has a unique jargon, that often is visible in the broader pattern, of how it structures its content.
One interesting topic to also talk about is not AI identifiers, but Human identifiers. Are there markers that genuinely can distinguish a human from an AI? I think this will increasingly get more attention as we compare human output to AI output and explore what the nature of the difference is really like.
Buzzwords, Tone, Inadequacy, and Exclusive Usage
Overuse of Buzzwords: Buzzwords are often seen as a sign of automation, but in business writing, they’re used by humans too. It’s a way to sound more professional or knowledgeable, so this isn’t a reliable indicator.
Inconsistent Tone: Some argue that automated content has an inconsistent tone. However, tone naturally shifts depending on the audience and context, something both humans and automated systems adjust to.
Doesn’t Answer Fully: There’s a belief that machines fail to provide complete answers. In practice, machines often provide more detail than needed, while human writers tend to be more concise, especially in marketing.
Unusual Word Choices: The use of uncommon words isn’t exclusive to machines. Both automated systems and humans can choose words creatively to make content stand out.
Spencer Romenco
Chief Growth Strategist, Growth Spurt
The BoostMyDomain team thanks these experts and leaders for taking the time to share their valuable insights on AI identifiers.
BoostMyDomain invites you to share your insights and contribute to our authoritative publication. Reach a wider audience, build your credibility, and establish yourself as a thought leader in an industry that caters to every business with an online presence!
