Learning to spot "AI slop"

I've been lucky enough to be included in three comprehensive training courses with a leading AI company - Mercor - on learning to spot "AI slop"! This has involved around 8 hours of solid training which has been really interesting.

OK so let's face it - the internet has always been full of low-quality content, so AI hasn't exactly started that. However, AI has made it possible to produce that content at an incredible scale. “AI slop” is the informal term for mass produced, low value content generated or heavily assisted by AI. It can take the form of articles, social media posts, videos, images, reviews and even entire websites. The tricky part is that AI slop doesn't necessarily look bad. It can be grammatically perfect, neatly structured and confidently written while saying almost nothing. Also, from my experience, AI can be absolutely terrible at producing images! Like so:

[Prompt was: Make the uploaded photo of me into a confused AI trainer at a computer with a cup of coffee on my desk] - not sure why I have three mugs of spilling coffee, a floating pen, a "reserch plan" and a clock that makes no sense!

One of the easiest ways to spot it is to look for vagueness and repetition. AI slop often uses lots of words without providing much useful information, relying on phrases such as “in today's ever-changing world” or “plays a pivotal role.” It may also repeat the same point several times using slightly different wording. The writing can feel suspiciously balanced, with predictable sections covering “the benefits,” “the challenges” and “the future,” without offering a strong or original argument. If you can remove half a paragraph without losing any information, that's a pretty good sign that you're reading fluff.

Another warning sign is a lack of specificity and personality. Human writers tend to include unusual observations, strong opinions, personal experiences or details that make their writing distinctive. AI slop often sticks to generic examples that could be used in almost any article. Instead of telling you what actually happened, it might say that a city has “vibrant culture and beautiful architecture” or that a technology offers “exciting opportunities and potential challenges.” These statements aren't necessarily false, but they're so general that they could apply to almost anything. Like a horoscope, I suppose!

Ultimately, spotting AI slop isn't really about becoming an AI detective; it's about becoming a better reader. Ask yourself whether a piece contains specific details, useful information, a genuine point of view, a reliable source and anything genuinely original. If it fails all five tests, it probably isn't worth your time, regardless of whether a human, an AI or both were responsible for writing it.