We live in a world where technology is rapidly advancing. You may have seen AI videos or writing on social media. During a time like ours, it’s important for everyone to have media literacy. By the end of this article, I’m confident that you’ll be able to spot AI-generated text anywhere. Oh, and one of these paragraphs was written by ChatGPT. Can you spot which one?

Why is it important to know when something is AI?

AI chatbots such as ChatGPT are known to actively spread misinformation. Although newer models of ChatGPT have begun showing sources for their information, they often make up their own facts without any. I’m a person who enjoys putting video essays on in the background while I do my homework or draw, but whenever I notice that there’s a sign of AI writing in the video, I immediately put on something else because I don’t want to support creators who are spreading misinformation.

Another reason AI spotting is an important skill to have is it allows you to support real journalists whose livelihoods depend on writing these articles. People who use AI as a shortcut are actively hurting the journalism industry, and they’re also creating content farms in the process. For example, on YouTube there are faceless AI channels such as Movies Explained that are actively hurting the platform by generating content farms that steal viewers away from youtubers who might spend weeks creating in-depth analysis.

Sign #1 - overuse of em-dashes

The first sign you’re reading something that was written by AI is the overuse of the em-dash, or this symbol: “—.” For whatever reason, AI loves using this symbol, but it's actually pretty hard to find on a normal keyboard. Writing this in Google Docs, I have to click the - button 3 times, which isn’t very obvious at all. Because this is such a strange symbol to be using so often, it's very rare to see anyone but AI using this. That being said, humans sometimes use this to sound more formal when writing, that’s why it's important to look out for other signs as well.

Sign #2 - “It’s not just _____; it’s _____.”

This is the most telltale sign for me personally. This is probably one of the most overused sentence formats in AI currently, to the point where it can even get obnoxious. Here’s a good example of Google Gemini using this format: “The strategy was not just comprehensive; it was also innovative, and most importantly, it was effective.”

Sign #3 - groups of 3

Another good sign of AI writing is the overuse of groups of 3. For example, AI might use 3 fun facts, or maybe 3 reasons why you should do something. Although humans also like using groups of 3, be sure to watch out for an overuse of them.

Sign #4 - meaningless metaphors

One way to spot AI-generated writing is exactly that: overcooked metaphors that feel like they’re trying to sound deep or poetic but end up sounding off, forced, or oddly generic. Here’s how that plays out and why it happens:

🔍 Why AI-Generated Writing Uses Weird Metaphors

Language models are trained to mimic human writing styles, and metaphors are a common stylistic tool. But without true understanding, they sometimes:

• Overuse metaphors to “spice up” otherwise plain content.

• Combine unrelated concepts that sound vaguely poetic but don’t actually make sense together.

• Miss cultural or emotional nuance, which humans intuitively pick up on.

🚩 Examples of Pointless AI-ish Metaphors

These are dead giveaways:

• “Like a lonely lighthouse in a sea of forgotten data…” — what?

• “The code danced like a ballerina on a stage of innovation.” — sounds dramatic, but… what is this even saying?

• “The heart of the economy beats in the chest of global commerce.” — huh?

They sound deep at first glance, but when you slow down, they’re hollow — all vibe, no meaning.

✅ How Human Writers Differ

Humans usually:

• Use metaphors sparingly and contextually.

• Tie them directly to the point they’re making.

• Drop them entirely if they don’t add clarity or impact.

Humans also revise — if a metaphor falls flat, a person usually notices it during editing. An AI, not so much (unless prompted to).

🧠 Tip for Spotting AI Writing:

When you read a metaphor in an article, ask yourself:

• Does this actually help explain something?\

• Does it relate clearly to the topic?

• Would a person naturally say this?

If the answer’s no — and especially if you see multiple metaphors like that — you’re probably reading AI-generated content.

Sign #5 - overuse of bullet points

Another sign that what you’re reading is AI is the repetitive use of bullet points. For whatever reason, AI loves using bullet points, even when they’re unnecessary. A great example of this is if you google something and you get the little AI-generated summary pop-up, the chances are you’re going to be seeing lots of bullet points. Here’s a great example of bullet points: