When Social Media Stops Feeling Real: How AI Slop Is Reshaping Online Life

Scroll through almost any major social media platform today and something feels different. Feeds that once mixed personal updates, news, and carefully made creative work are increasingly filled with strange images, repetitive videos, and emotionally charged scenes that feel artificial, exaggerated, or simply nonsensical.

This growing wave of low-effort, AI-generated content has become known online as “AI slop”. It is not a technical term, but it captures a shared frustration. Content that is cheap to produce, designed for fast emotional reaction, and optimised for engagement rather than meaning.

What began as novelty has quietly turned into saturation, and many users are beginning to push back.

What people mean when they say “AI slop”

AI slop usually refers to images and videos generated quickly using artificial intelligence tools, often with little care for realism, coherence, or ethics. Common examples include fake images of children in distress, miraculous acts staged for sympathy, animals in improbable danger, or surreal religious and military scenes designed to provoke emotion.

The aim is not accuracy or storytelling. The aim is reaction. Likes, shares, comments, and watch time.

Because modern algorithms reward engagement above all else, this type of content spreads easily. It requires no filming, no editing skills, and no real-world accountability. A single creator can generate dozens of posts a day, testing which ones trigger the strongest response.

Why platforms quietly benefit from the flood

Major platforms have not resisted this trend. In many cases, they have encouraged it.

Companies like Meta and Google have openly described artificial intelligence as the next phase of social media. Built-in image generators, video tools, and AI filters are now standard features, making content creation faster and more accessible than ever.

From a business perspective, AI slop is efficient. It keeps users scrolling, costs very little to host, and scales infinitely. Whether the content is meaningful is largely irrelevant to the system that distributes it.

Research into platform feeds suggests that a noticeable proportion of content shown to new users is already low-quality AI-generated media, particularly in short-form video formats where speed matters more than depth.

The growing sense of backlash

While AI slop performs well numerically, sentiment around it is shifting.

Under many viral posts, the most visible comments are no longer admiration but irritation. Users point out obvious flaws, complain about deception, or express exhaustion at constantly having to question what is real.

In some cases, comments criticising the content receive more engagement than the content itself. This creates a strange feedback loop where outrage still fuels visibility, further embedding the very material people want less of.

A small but notable part of this backlash has taken shape through online accounts dedicated to highlighting absurd or manipulative AI-generated posts. One such account, run by a young student in France, catalogues extreme examples of AI slop circulating on platforms like Facebook. The account has drawn attention to how easily such content gains traction without scrutiny. You can find it here: https://x.com/FacebookAIslop

The existence of accounts like this reflects a wider mood rather than a single campaign. A sense that something about the online environment is slipping out of balance.

The mental toll of constant artificiality

Researchers studying online behaviour warn that the impact of AI slop is not just annoyance.

Constant exposure to content that is fake, exaggerated, or meaningless can reduce attention span and discourage critical thinking. Verifying authenticity requires effort. Over time, many users simply stop checking.

This has led some academics to describe a “brain rot” effect. Not because individual videos are harmful, but because the overall environment trains people to consume quickly, react emotionally, and move on without reflection.

Even content that is obviously fake can contribute to this erosion by normalising a feed where nothing needs to make sense to succeed.

When slop turns into something more serious

Beyond irritation, AI-generated content can carry real risks.

Recent controversies involving AI tools being used to digitally alter images of real people, including women and children, show how quickly low-quality content can cross into abuse. In other cases, fake videos and images have been used to shape political narratives, creating the illusion of public support or emotional response that may not exist.

This is especially concerning as many people now rely on social media as their primary source of news and information.

At the same time, several major platforms have reduced human moderation, relying more heavily on automated systems and user reporting. This makes it harder to respond quickly or consistently to emerging harms.

Where this leaves us

AI-generated content is not going away. The tools are improving, the costs are falling, and platforms remain financially aligned with volume over quality.

The question is not whether AI will be part of online culture, but whether digital spaces can retain any sense of trust, creativity, or shared reality if everything becomes synthetic, disposable, and engagement-driven.

For many users, the frustration is not about technology itself, but about what it is being used for. The fear is not of AI creativity, but of an internet increasingly filled with noise, manipulation, and content designed to exploit attention rather than inform or inspire.

If there is a shift coming, it will likely come not from platforms, but from users deciding what they are willing to tolerate in their feeds, and what they quietly stop engaging with.

*All images generated on Leonardo AI