AI-assisted writing tools leave fingerprints. Your readers might not be able to name what bothers them. They'll just feel it. Something slightly off. A little too smooth. A tad too predictive. A bit too even. And then they'll quietly scroll past your content (AI;DR).

If you work in communications, public relations, content creation, public affairs, marketing, or any field where your brain/creativity/strategy is your currency, this is actually incredible news. It means that we humans with decades of experience and expertise under our belts may just survive Altman's dystopian "Gentle Singularity." :)

And just like gamified social media algorithms a few years ago, it's now time we learn (and stay on top of!) the dead giveaways that artificial intelligence wrote your content. Not only to keep our audiences engaged and avoid creating more slop, but also to build another necessary tool in our professional kit.

I pored through listicles, blogs, and YouTubes to bring my favorite top 21 AI writing tells. Most are fixable with a good prompt and a careful edit (your authentic voice (or organizational brand voice) must still carry the day!). Friendly disclaimer: I'm also not saying to never use anything on this list, just be aware and choose intentionally.

And yes, I then had Claude AI synthesize my research, notes, thoughts, and ideas. Please enjoy and share your feedback to laura@onmessage.co.

1. The ai-diolect cluster

AI models dramatically overuse a recognizable set of formal, slightly literary words that almost never appear in natural professional writing. The full red-flag list includes: delve, tapestry, realm,

beacon, nuanced, robust, multifaceted, intricate, cornerstone, harness, foster, illuminate, paramount, leverage, and navigate (used metaphorically). Expect this list to shift and evolve.

    AI: "This initiative underscores the pivotal role of harnessing innovative strategies to navigate today's multifaceted digital landscape."

    Human: "This project shows why we need new tactics for online marketing."

2. Significance inflation

AI inflates the importance of ordinary topics with grandiose language. A regional conference becomes "a groundbreaking testament to the transformative power of innovation." A software update becomes "a paradigm shift in how organizations approach workflow optimization." Telltale intensifiers include pivotal, crucial, vital, revolutionary, game-changing, unprecedented, and cutting-edge. When everything is extraordinary, nothing is.

3. Em dash overload

This was the single most discussed AI punctuation tell of 2024 and 2025. By mid-2025, more than half of ChatGPT's responses contained em dashes, up from fewer than 1 in 10 the year before. The Washington Post, NPR, and Rolling Stone all covered the phenomenon. Em dashes are legitimate punctuation, but AI deploys them with mechanical regularity as a universal connector for emphasis, pivots, and parenthetical asides. Three or more in a single paragraph is a strong signal. Some editors have started calling it "the ChatGPT dash."

This one is particular sad for me, as I heart the em dash and miss it terribly. #rip

4. "It's not X, it's Y" Sentence Construction

This is AI's favorite rhetorical device. The negation-then-assertion construction shows up constantly in AI-assisted LinkedIn posts, blog intros, and marketing copy. "We're not just building a product, we're creating an experience." Followed three paragraphs later by: "It's not about working harder, it's about working smarter." AI repeats this formula without registering that it has already used it. Two or more in one piece is a clear red flag.

5. The "In today's world" opener

AI consistently opens content with generic temporal framings that could apply to any topic in any era. These openers attempt context-setting but add zero actual information. If your content starts with "In today's rapidly evolving landscape," delete the first sentence and start with whatever came second.

6. The rigid intro-body-conclusion sandwich

AI applies the same mechanical structure to every piece regardless of format: broad contextual opener, organized middle sections, tidy summary conclusion. Human writers vary structure based on purpose. Some pieces start mid-story. Some skip a formal conclusion entirely. When every piece of content from the same source follows the exact same arc, the machine origin shows.

7. Uniform paragraph length

Researchers measure sentence variation as 'burstiness,' and studies show AI text scores significantly lower than human writing on this metric — meaning AI paragraphs tend toward uniform, metronomic sentence length while human writing naturally mixes short punchy lines with longer, winding ones. Human writing alternates naturally between short, punchy lines and longer, winding passages. One-sentence paragraphs are fine. So are longer ones that unpack a single idea across eight sentences. Metronomic uniformity throughout a piece is a tell.

8. The rule-of-three default

AI consistently groups items in threes. Three examples. Three benefits. Three takeaways. Two feels incomplete to a language model; four feels excessive. Read back through a few AI-generated pieces and count the groupings. The pattern becomes obvious fast.

This one also bums me out, so I'm now heading the opposite direction with five+ groupings, especially adjectives. :)

9. Excessive headers and bold-keyword formatting

AI defaults to Title Case headings, dense subheaders, and a distinctive "Bold Term: explanation" inline format that rarely appears in organic professional writing. It treats every section as requiring a formal heading, even in short pieces. The result looks like a Wikipedia article when a memo would serve better.

10. Transition word overload

Researchers at Carnegie Mellon found that some AI-favored words appear up to 150 times more often in AI outputs than in comparable human writing. 'Furthermore' and 'moreover' show the same pattern, technically correct, but never how anyone actually talks.

11. "It's important to note that..."

It sounds authoritative but adds nothing. Variants include "It's worth noting," "It should be noted that," and "It bears mentioning." Delete every instance and read the sentence without it. It always holds up fine on its own.

It's important to note that this one also bums me out, as I love to note important things.

12. Excessive hedging and qualification

Beyond individual phrases, AI has a systemic pattern of over-qualification. Every claim gets softened with multiple qualifiers stacked together: "Generally speaking, in many cases, content marketing can potentially be an effective strategy, though results may vary depending on a variety of contextual factors." Research confirmed that ChatGPT significantly increases hedges and reduces boosters. That's the exact opposite of persuasive professional writing.

13. Avoidance of contractions

AI defaults to "it is," "you will," and "do not" rather than "it's," "you'll," and "don't," even in contexts that call for warmth and a conversational register. Marketing copy, emails, and social content written without contractions feel oddly formal. "You will appreciate the comfort this home provides" is not how anyone actually talks.

14. Lack of genuine personal voice

AI text defaults to a formal, emotionally detached tone. It can't draw on real experiences, name specific colleagues, recall sensory details, or express genuine frustration, doubt, or surprise. It reads like a statistical average of a million voices: polished and soulless. Readers feel this even when they can't articulate it.

15. Single grammatical voice with no natural shifts

AI locks into one grammatical person and holds it rigidly: all second-person "you," or all third-person "they." Human writers naturally move between first, second, and third person when telling stories, injecting personal reactions, or quoting sources. An article that stays entirely in second person for 1,500 words carries a mechanical quality that experienced readers pick up on.

16. Hallucinated statistics

AI fabricates statistics, citations, case studies, and quotes and presents them with full confidence. In one study, 43.5% of surveyed marketers reported that completely false AI-generated information made it past review and went live publicly. Every AI-generated data point needs independent verification before it reaches a client or a public audience.

17. Vague attribution

AI cites authority without providing actual sources: "Research indicates," "experts agree," or "a recent survey found." This creates the appearance of evidence-based writing while providing nothing verifiable. Any claim that begins with one of those phrases and doesn't end with an actual citation needs to be cut or sourced.

18. Synonym cycling

AI has been trained to avoid word repetition, so it cycles through synonyms unnecessarily. The same entity gets different names within a single passage: "The protagonist faces many challenges. The main character must overcome obstacles. The central figure eventually triumphs. The hero returns home." Pick one term and use it. Repeating a proper noun is not a writing error. Synonym cycling is.

19. Bullet point addiction

AI immediately defaults to numbered or bulleted lists, even for complex topics that deserve flowing prose. It also decorates lists with emoji: checkmarks, brain icons, sparkles. Humans discuss nuanced ideas in paragraphs and use lists selectively. Asked "How can I be happier?", a language model gives you five bullet points and a trophy emoji. A human writer gives you a paragraph that starts with an honest observation.

20. Unnecessary term definitions for expert audiences

AI defines terms that professional readers already know. In a post written for marketing directors: "Search Engine Optimization, commonly known as SEO, is the process of improving your website's visibility in search engine results pages (SERPs)." If you're writing for a professional audience, they already know what SEO stands for. Calibrate to the room.

21. Each model leaves distinct fingerprints

ChatGPT, Claude, and Gemini each have recognizable stylistic signatures. ChatGPT leans heaviest on em dashes, bold formatting, bullet structures, and words like "certainly" and "utilize." Claude writes more flowing prose with less formatting, favors measured hedging and introspective tone, and reaches for phrases like "it appears" and "it's worth considering." Gemini produces utilitarian, report-like output with more frequent italics and the phrase "not only... but also."

When multiple writers on a team use the same model without heavy editing, their output starts to converge into suspiciously uniform prose. That's a team-level tell, and it's increasingly recognizable to editors, journalists, and anyone who reads a lot of professional content.

The one thing no prompt can fix is a real voice. That still requires a real person. Your audience can tell the difference, even when they're not trying to.