Slop goes the weasel - how trashy AI content is killing user interest

AI-generated "slop" is flooding the internet even as engagement plummets. Facebook engagement has dropped 36%, Instagram 16%, TikTok 34%, and X 48% year-on-year. Meanwhile, 39% of people actively avoid news due to overwhelming volume. The winning strategy? Post less, focus on quality, and keep humans in control.

---

Garbage in, garbage out. The internet is filling up with AI-generated “slop”. We're drowning in a rising tide of low-effort, synthetic posts that are designed to feed algorithms rather than inform or delight us.

At the same time, user behaviour is showing growing fatigue, falling engagement, and increasing anxiety about what is real, suggesting people are starting to tune out from the flood rather than swim in it.

Is there anything creators can do to counteract this alarming race to the bottom?

What is “AI slop”?

“AI slop” is the catch-all label for the mass of derivative listicles, bland blog posts, generic LinkedIn threads, spammy product reviews and uncanny images pumped out by AI tools with minimal human editing. Built by machines to garner any trace of engagement, no matter how fleeting, slop is optimised for volume, keywords and ad inventory, not for insight or originality.

Generative tools now promise to auto-write blogs, captions and emails, and to batch-schedule content across every major platform from a single dashboard. This dramatically lowers the marginal cost of “one more post”. As one 2025 guide for marketers put it bluntly: Brands using AI in their workflows can pump out around 40–45% more content per month while cutting production costs by about a third.

This shift fundamentally changes the economics of attention: Content supply is exploding, but human attention (demand) is not.

How fast is the slop quotient growing?

One way to measure the “slop quotient” is not to look at how many posts are being published, but at what is happening to engagement per post. Across almost every major social network, that metric is heading in the wrong direction.

A 2025 cross-industry benchmark of social media performance shows that engagement rates fell on every major platform year-on-year:

• Facebook’s median engagement dropped 36%

• Instagram’s fell 16%

• TikTok’s plunged 34%

• X suffered the steepest fall at a stunning 48%

At the same time, many industries either maintained or increased posting frequency, creating more content that is fighting for fewer and fewer interactions.

Separate analysis of 125 million posts found Instagram’s engagement rate dropped by about 28% year-on-year, even though it still outperforms Facebook. In other words, feeds are getting fuller while individual posts are drawing less response. This is exactly what you’d expect when automation makes it trivial to fill every slot in the calendar.

Unfortunately, AI vendors are leaning hard into this dynamic, chasing revenue from every type of business. Content automation platforms now advertise the ability to “create once, distribute everywhere,” and to spin up dozens of variations of the same core asset for different channels, audiences and all times of day.

Even if the headline claims about “hundreds of posts per hour” are more marketing bravado than standard practice, the direction of travel is clear: The marginal cost of pushing out another near-identical post is approaching zero.

Are users switching off? The engagement and avoidance data

If slop is rising, the key question is whether audiences are happily consuming it or are they quietly walking away?

The evidence points firmly to the latter.

On social platforms, declining engagement rates already suggest users are interacting less with each individual post, even as their feed fills up. Some sectors, such as retail and fashion, are seeing double-digit drops in engagement despite intense posting, indicating that simply “posting more” is no longer an effective strategy.

Zooming out to news and information, the pattern is even more stark:

The 2024 Digital News Report from the Reuters Institute found that around 39% of people across 20 key markets now say they sometimes or often avoid the news, up 3% in a single year and the highest level since tracking began. In the UK, the share of people who say they avoid news sometimes or often jumped from 24% in 2017 to 46% in 2024.

When asked why they avoid news, people cite the overwhelming volume of content and the negative impact on their mood, alongside a sense that much of what they see is repetitive or unhelpful. Those answers dovetail uncomfortably well with a feed stuffed with AI-spun rewrites, outrage-bait headlines and low-quality commentary.

Put more bluntly: As the slop quotient rises, a growing number of users choose to look away.

Trust, fakery and the feeling of being tricked

The other side of the slop story is trust. As generative tools make it easier to fabricate plausible-looking text, images and video at scale, audiences are becoming less sure of what they see.

Belief is being tested.

Globally, 59% of surveyed people said they worry about what is real and what is fake when it comes to online news, up from 56% the previous year and around five percentage points higher than two years ago. On social media specifically, around two-thirds of US consumers believe that more than three-quarters of the news they see is biased or even fake.​

As AI-generated slop floods search results and social feeds, users report feeling both overloaded and manipulated. They complain about clickbait rewrites of the same wire story, bizarre AI-generated imagery used to illustrate tragedies, and articles that read like stitched-together summaries rather than genuine reporting. Taken together, these experiences erode confidence that what users see is real.

Are people really shunning slop, or are they just getting used to it?

The data points to a nuanced picture. People are clearly tired of endless, low-quality content, but they are not abandoning digital platforms altogether. Instead, they're quietly changing how they use them.

First, there is evidence of a shift away from generic feeds and towards more controlled, interest-driven environments. The Reuters report notes that going directly to news websites and apps as a primary gateway has fallen by about 10% since 2018, while platforms like YouTube, TikTok and Instagram have become more important routes to news - especially for younger users.​

Second, users are increasingly navigating around slop rather than quitting platforms altogether. They mute or unfollow low-quality pages, subscribe to specific newsletters and podcasts, and build lists of trusted creators instead of leaving discovery entirely to the algorithm. This is a kind of “micro-avoidance”: Even if they stay on the platform, users try to fence off the worst of the feed.

Third, when people talk about what they want from news and information, they consistently ask for the opposite of slop. In open responses, audiences say they would like fewer constant updates and more context, more explanation, more personal utility and more stories that help them understand a complex world. Most say they do not want the news to be “more entertaining”, they want it to be more useful and hopeful. That preference is hard to reconcile with an environment optimised for relentless, automated content output.​

Escaping the slop trap – less is more

For creators, brands and publishers, the clear message is that volume-led AI strategies are rapidly hitting diminishing returns. The data suggests the winning move is not to flood the zone, but to refocus on distinctiveness and human judgment.

Social media benchmarks show that sectors such as higher education and some non-profits still outperform average engagement with relatively modest posting volumes. This suggests that relevant, community-rooted content cuts through even in saturated feeds. By contrast, media accounts that blast out vast numbers of posts see very low engagement per piece.​

Similarly, marketing analysis indicates that brands which post less frequently but invest in higher-quality creative often see better engagement than those that treat AI as a firehose. Algorithms tend to de-prioritise accounts that consistently fail to elicit responses, so spamming your followers with slop can be actively self-defeating.

The emerging best practice is to use AI as a partner rather than a replacement: To brainstorm, draft, repurpose and translate, but to keep humans firmly “in the driving seat” for story choice, framing, nuance and verification. That approach aligns with audience preferences, which are far more accepting of AI working behind the scenes rather than AI writing entire articles or fabricating images on sensitive topics such as politics and war.

Lastly, stop counting engagements, impressions and likes as success. When 40 different versions of the same content bombard millions of feeds but nobody watches or reads what you post, the volume is meaningless not meaningful. Instead, concentrate on building real, lasting relationships with an audience that actually cares about what you say. 1,000 viewers who take the time to invest in your commentary, thoughts and brand and far more valuable than 100,000 who simply scroll past your screed.

Final word:

In an AI-saturated landscape, the rarest resource is not content but discernment. The winners will not be those who can generate the most posts per hour, but those who can decide what not to publish, and who can earn enough trust that, so that when they do speak, people choose to listen.

---

Get started with Freelance Words

Strong communication is the bedrock of good business and when there’s a lack of it, problems can arise. Take the ambiguity and doubt out of what your business needs to say. Contact us now.

Contact