Think You Can Spot AI Writing? Here’s How Our Editors Do It

These days, distinguishing AI writing from human writing is like telling apart identical twins. It is tricky, and sometimes impossible, without a careful eye. 

As a writer, I’ve stared down blank pages, wrestled with stubborn sentences, and felt the unique thrill of finding the perfect phrase. Writing is a human craft, messy, intuitive, and deeply personal. 

However, lately, I have found myself playing a new kind of game: "Author or Algorithm?" 

It always starts innocently. When reading a product description, a blog post, or a LinkedIn article, the grammar is flawless. The sentences are structured correctly. The information seems accurate. But something feels… off. A faint, digital hum lies beneath the words; a tell-tale sign that you might be reading the work of artificial intelligence. 

We, editors and content writers at REM Web Solutions and other leading publications, use a blend of technical insights, writerly intuition, and hands-on practice to separate AI-generated content from authentic human writing. 

So, do you think you can spot AI writing? Let’s pull back the curtains. Here are the top clues we at REM Web Solutions look for, and how you can train your own eyes to see them. 

Why it Matters 

Before diving into the clues, let’s talk about why this matters. AI writing is no longer sci-fi. Models like ChatGPT, Claude, Gemini, and others can generate full articles, marketing copy, social posts, and even poetry in seconds. The benefits are obvious: speed, scale, and convenience. But that power also brings risks: 

• Misinformation, errors, and hallucinations. AI can invent facts, sources, or misstate something with confidence. 
• Lack of voice or uniqueness. AI tends to flatten nuance, emotion, and personal style. 
• Ethical and transparency concerns. Readers deserve to know whether they’re engaging with human or machine writing. 
• Reputational risk. If your content reads “robotic” or “promo fluff,” your credibility suffers. 

So, whether you're an editor, content writer, educator, or curious reader, knowing how to spot AI writing is a critical skill. 

Red Flags: Telltale Signs of AI Writing 

1. Repetitive Patterns and Overused Phrases 

AI models frequently recycle sentence structures or use the same phrasing throughout a piece. This can be repeated introductory lines, summaries, or catchphrases. For example, you might notice the phrase “It’s important to note that…” appearing in multiple paragraphs is an indicator that the text may have been generated by an AI. 

2. Monotonous Rhythm & Flat Tone 

Human writing always shifts naturally between different lengths and styles of sentences, showing “bursts” of energy. Human writing also highlights ideas with punctuation or creative turns of phrases. Whereas AI writing is more uniform, paragraphs have a robotic rhythm, and the “voice” sounds polite, repetitive, or bland. Humans, by contrast, naturally break rules for emphasis and style. 

3. Excessive Use of Em Dashes 

AI is a fan of the em dash (—). AI models disproportionately use the em dash (—). While em dashes are a standard part of English punctuation, their repetitive and unnatural placement is a common red flag feature in AI-generated content. Most human writers use them sparingly, while AI text sometimes substitutes dashes for other punctuation conventions at odd or inappropriate times. 

4. Generic and Vague Content 

Look for language that is broad, generic, and lacking in concrete detail or personal perspective. AI-written text will often “sand down” unique facts, ending up with bland generalizations. You’ll see descriptions that could fit almost any topic and lack the specifics that tie the writing to real experiences or genuine expertise. 

5. “Servile Positivity” and Overly Polite Statements 

Most AI tools are programmed for helpfulness and safety, resulting in language that is overly positive and submissive. Human writers are more likely to include bold opinions, disagreements, or creative risk-taking that adds texture to content. 

6. Sudden Changes in Quality or Style 

Because some bloggers and companies blend human and AI content, a single article may abruptly shift in complexity, depth, or vocabulary. If one section suddenly becomes more technical or overly wordy, or the tone becomes oddly formal or simplistic, it’s a good indication of AI involvement. 

7. Flawless Grammar 

While skilled writers can produce clean copy, humans often have minor errors, intentional grammar breaks, or stylistic quirks. AI-generated content tends toward perfect grammar, which can come off as uncanny or sterile in larger volumes. Editors sometimes notice a sudden disappearance of slang, voice, or regional expressions (such as Canadian-specific language). 

8. Incorrect or Invented Facts 

AI models can “hallucinate” facts by either inserting believable but incorrect information or by creating fake citations and sources. Editors should always cross-check references and verify claims, flagging content that sounds authoritative yet can’t be substantiated. 

What Makes Human Writing Stand Out 

Our editors don’t just look for signs of AI; they also spot what makes writing truly human. 

• Personal touch: Stories or ideas based on real experiences. 
• Thoughtful insight: Looking deeper and making connections that matter to readers. 
• Varied style: Changing sentence length and tone to match the feeling or purpose. 
• Emotion and humour: Small hints of wit, subtlety, or emotion that show real understanding. 

Tools and Techniques Editors Use 

Our editorial teams (content writers) at REM Web Solutions blend manual reading with technology for optimal accuracy. 

Manual Review 

• Read aloud to “hear” if the text sounds natural. 
• Look for the author’s voice (personal anecdotes, opinions, or direct addresses) to the reader to suggest a human touch. 
• Check for inconsistencies: sudden changes in writing style, unexplained jumps in logic, or strange repetition signal possible AI use.  

AI Detection Tools 

• Use AI detectors like ZeroGPT, GPTZero, or Writer AI Detector to scan for machine-generated patterns. However, keep in mind that these tools can also fail to properly detect AI patterns and may even claim human-written content is AI, as AI borrows from existing written material that is available online. Like AI, they are just a tool. 

Fact-Checking and Cross-Referencing 

• Always verify unusual facts, direct quotations, and industry stats. 
• Watch for invented references or links that don’t point to credible sources. 

Common Myths and Limitations 

1. No single tip or tool can catch all AI writing. Detection is about context and cumulative signals.  
2. As tools improve, AI text becomes more nuanced and harder to spot, so human expertise matters more than ever.  
3. Human writers can sometimes appear “robotic” (especially with templates), so avoid jumping to conclusions without thorough review.  

Beyond the Words: How to Identify Who's Posting AI-Written Content 

Spotting the writing itself is one thing. But how can you tell if a person or company is routinely relying on AI without human oversight? Our team is looking for patterns. 

1. Superhuman Output: Does a single author or company publish dozens of long, "well-researched" articles per week? While not impossible, a consistent volume of high-quality, in-depth content is a major red flag. Quality human writing takes time. 
2. Generic Social Media Posts: Check out their LinkedIn or Facebook feeds. Are the posts that promote the article also bland and formulaic? ("Check out our new article on X! It covers important tips for Y.") A human promoter would add a personal hook or a key takeaway. 
3. Lack of Engagement: Read the comments on their blog or social media posts. Is the author responding? Are the answers generic? AI-written content rarely sparks genuine conversation because it lacks a point of view to engage with. 

The Human Touch: What AI Can’t Replicate (Yet) 

AI can mimic writing, but there’s something special that only humans bring. 

1. Personal Stories: Sharing a client's experience, a funny mistake, or a lesson you learned the hard way. AI can’t create this in real life. 
2. Strong or Thoughtful Opinions: Taking a clear stance on a tricky topic or exploring the grey areas that AI usually oversimplifies. 
3. Cultural and Contextual Humour: Jokes, sarcasm, and wit that depend on shared understanding of current events or local culture. 
4. Emotional Impact: Writing that truly makes readers feel something—like joy, empathy, or inspiration. AI can describe emotions, but it can’t create them from real experience. 
5. Interview Insights: Quoting a real conversation, capturing the speaker’s unique voice and personality. 

Conclusion: Why Human Editors Matter Now More Than Ever 

AI can streamline content production and offer new tools to marketers, but only experienced editors can ensure authentic, relevant, and reliable writing that reflects real perspectives. 

The key is the human in the driver’s seat. A skilled editor or writer uses the AI-generated draft as raw clay, not a finished sculpture. They inject the voice, the nuance, the original insights, and the factual verification. They turn the generic into genuine. 

At REM Web Solutions, our editorial expertise makes the difference, so that our clients, partners, and audiences always know who’s really behind the words. 

Ready to ensure your web content stands out for all the right reasons? Chat with our team and bring an expert’s eye to your AI-era digital strategy.