Online content writers are perhaps the biggest contributors to the wave of generative AI text that is currently flooding the internet.
They’ve figured out how to publish articles on an industrial scale, and although it’s hard to say whether these tactics are working for them, it’s safe to say the quality of the content leaves a lot to be desired.
You can tell when someone is putting out automated content by looking closely at the fluff and filler that pervades AI generated articles.
Indeed, LLMs such as ChatGPT have a distinct set of traits when it comes to padding out a piece of text—we’ve listed 5 examples to help you become a pro at identifying spammy AI content.
1. Content with spammy AI titles
It’s always worth making an effort to keep up with the tell-tale signs of titles written by AI, because then you can dismiss the content without first having to click on it.
Watch out for bold adjectives in titles for instructional content such as guides:
The Ultimate Guide to SEO
AI knows that a guide titled in this way will appeal to a lot of readers because it suggests all the information you could ever need is conveniently packaged together in one place.
But of course, long-form content isn’t AI’s strong suit and so when users visit the page they are invariably disappointed to find out the content is scant at best.
Also keep an eye out for titles which are verbose and often redundant:
The Ultimate Guide to SEO: Mastering Search Engine Optimization
The second part of the title is essentially a repeat of the first, adding little to no SEO value. If you see a title containing a colon, this should put you on high alert for possible artificial intelligence.
2. Content with spammy AI words
Everyone knows that keyword stuffing is bad SEO practice, but AI chatbots must have missed that memo because they’ll happily insert target keywords left, right and center.
And you can forget about keyword variation—ChatGPT doesn’t have the processing power to realize that using related terms in different locations (including the meta description) does wonders for capturing a wider range of search queries.
Keywords aside, AI has a list of favorite words it likes to employ in all its content!
For example, elevate has a habit of appearing in titles, headings, and the main body, and we mean in the same piece of content. It’s so noticeable that you can probably detect AI content on the basis of how many times ‘elevate’ is used.
3. Content with spammy hyphenated words
The phenomenon of ChatGPT hyphenated words is such that it deserves to be discussed separately from individual words.
For certain topics such as technology, business or sociology, which many people would describe as rather dynamic topics, AI-generated articles tend to include hyphenated words in the introduction.
- Ever-changing
- Ever-evolving
- Ever-expanding
- Ever-competitive
- Hyper-connected
- Fast-paced
- Digitally-driven
It’s theorized that AI chatbots like using dynamic language in the form of compound adjectives as it better reflects the dynamic nature of certain topics.
That’s all well and good, but when you employ them as routinely as AI does, and in the exact same location in every article, you end up with a pile of content that resembles automated AI spam.
4. Content with spammy AI phrases
Yes, AI goes so far as utilizing an entire inventory of its favorite phrases to lengthen its content and meet a specified word count.
It’s one of the reasons why you should never prompt an AI chatbot to write more than what they are comfortable with, otherwise you will experience an overload of unnecessary words and phrases.
There are too many phrases to list here, so be sure to check out our top 10 list of AI phrases, and prepare to be blown away by unremarkable prose like “x is a rich tapestry of y and z” along with false promises of “embarking on a journey”.
5. Content with spammy AI images
It’s no surprise that AI images and AI art have become interchangeable terms—even AI-generated images which are supposed to look photorealistic have a cartoonish quality about them.
At least the unrealistic nature of this on-demand form of visual media makes it so much easier to spot them for what they are: spammy AI.
AI images are part and parcel of the generative AI movement that’s currently taking over the web, as more and more publishers seek to increase their online presence through automated content production.
Thankfully, AI images have a negative impact on SEO because it raises doubts over how much of an authority the author is within their niche.
Conclusion
The presence of certain words, phrases, and images can help you determine to a high degree of accuracy whether a piece of content is a product of AI.
If someone is publishing content with these tell-tale signs of AI on a mass scale then it qualifies as AI spam.