AI is filling the internet with slop. We must fight back.

Slop, the unsavoury term for the low-quality, error-filled content filling up online sources, is creeping close to epidemic proportions.

You can’t have missed it—it’s in images where you can see anatomical impossibilities like a cat with three tails. It’s in text, in which figures are all wrong, and the language is robotic and full of known patterns. It’s in art and music that have a false ring to them, in misleading videos on YouTube, social media, blogs, and forums.

It’s polluting the internet. Before we know it, there may be only less trustworthy, low-quality content left. Clearly, we had better fight back—now.

The frightening thing is that some AI tools may be taking in the content they generated for training. A vicious loop will set up, degrading content further unless there’s some human intervention built in to change the trajectory.

Quality over quantity

How did the AI slop situation get so bad so quickly? There’s a whole collection of reasons working together. At some point in our journey online, we began to prioritize numbers and volume over quality. We create content for SEO and look for enough numbers to generate advertising. We look for influencers to spread the content, even if it’s sometimes for bots to consume. The tools to create content became cheaper, quicker, and more powerful. Gradually, human evaluation of quality became distorted.

I remember crying for a week when a typo (little cell phone instead of little cellphone) escaped into print in the magazine of which I was the editor. Today, that reaction would be bizarre.

AI is far from perfect, being as young as it is. Errors, hallucinations, sycophancy, and more cloud the content it produces. AI doesn’t actually understand what it’s saying, so it can’t easily correct or fact-check itself. So, we have ourselves an avalanche of flawed content.

But there is good along with the bad, and since AI is here to stay for the foreseeable future, we have to deploy strategies to fight back against low-effort content.

Slop can sometimes be difficult to spot. You could watch a video and be halfway through before you get the sneaking feeling something isn’t right. You actually get slop detectors online. Feed in your text or image, and they give you an AI generation score.

Try SlopDetector, SlopOrNot or Slop Detect, for a lark and see what you get. I am deeply offended to find paragraphs from this article tagged as 92% AI-generated and having the “creative energy of a wet napkin". This is fun.

AI detectors try to catch the AI out on certain patterns, but the AI, in turn, is learning to humanize the content to escape detection and be more palatable. Rather than relying on doubtful detectors, the most important strategy is to always be alert where content is concerned.

Go old school

As always, check for reliable sources, obvious agendas, and attempts to sensationalize, shock, or draw out emotions. Slop and the ability to create content at speed and volume combine with the intention, by some creators, to mislead and influence or at least to persuade you to stay or keep coming back for more.

If you’re requesting content, make sure your prompts are clear, specific and detailed to minimize the chance of errors. There are many guides online to learn better prompting, but one interesting way is to ask the AI assistant when you do get a desired result: How could I have improved on this prompt.

Anti-slop strategies must include ensuring not to endorse poor content or to do anything to help spread it. On YouTube, for example, fine-tune your recommendations and choose not to get further content from a channel that seems to be low quality or has an agenda. In text, be watchful for repetition, generic and vague phrasing, and claims without examples. Look for known authors, quoted sources and references that signal actual work done. Do a reverse search—if something seems shallow, run a quick search to see if it’s available elsewhere.

We all create some content—ensure you produce quality. Use AI chatbots as research assistants rather than overly relying on them for content.

The New Normal: The world is at an inflexion point. Artificial intelligence (AI) is set to be as massive a revolution as the Internet has been. The option to just stay away from AI will not be available to most people, as all the tech we use takes the AI route. This column series introduces AI to the non-techie in an easy and relatable way, aiming to demystify and help a user to actually put the technology to good use in everyday life.

Mala Bhargava is most often described as a ‘veteran’ writer who has contributed to several publications in India since 1995. Her domain is personal tech, and she writes to simplify and demystify technology for a non-techie audience.