Social Media Feeds Fill with AI ‘Slop’ as Users Begin to Revolt

Théodore remembers the AI-generated content that finally pushed him over the edge.

The post showed two malnourished South Asian children. Despite their youth, both had thick beards. One had no hands and only one foot, while the other held a sign claiming it was his birthday and asking for likes. Strangely, they were sitting in the middle of a busy road in pouring rain, with a birthday cake in front of them. Every detail suggested the image was AI-created, yet it went viral on Facebook, attracting nearly a million likes and heart emojis.

“It blew my mind. These absurd AI images were everywhere, gaining massive attention without anyone questioning them—it was insane,” says the 20-year-old Paris student.

In response, Théodore created an X account called Insane AI Slop, highlighting and mocking misleading AI content. His inbox quickly filled with submissions from others who noticed similar patterns—religion, military, or impoverished children performing extraordinary acts.

“Kids in the developing world doing impressive things is always popular. People see it as wholesome, so creators think, ‘Great, let’s make more,’” he explains.

His account grew to over 133,000 followers.

The flood of AI slop—quickly made, low-quality, unconvincing images and videos—is now unstoppable. Social media platforms have embraced AI, and while some companies claim to be addressing low-quality content, feeds remain filled with it.

The rise of AI-driven social media

In a recent earnings call, Meta CEO Mark Zuckerberg described social media as entering a “third phase,” now dominated by AI.

“First, content came from friends and family. Then creator content was added. Now, AI makes it easy to generate and remix content, creating a massive new layer,” he said.

Meta, which owns Facebook, Instagram, and Threads, not only allows AI-generated content but has launched tools to create more of it, including image and video generators and advanced filters.

YouTube CEO Neal Mohan reported that in December 2026 alone, over a million channels used the platform’s AI tools. He acknowledged concerns about low-quality AI content and said efforts are underway to limit repetitive or poor-quality videos, though he avoided defining what should or shouldn’t thrive.

Research by AI company Kapwing found that 20% of content shown to a new YouTube account is low-quality AI video. Short-form videos are particularly dominated by it. Channels like India’s Bandar Apna Dost have amassed 2.07 billion views, earning creators an estimated $4 million annually.

Backlash emerges

At the same time, users are pushing back. Under viral AI videos, furious comments now appear frequently.

Théodore has been part of this movement. On X, he reported bizarre and disturbing AI videos to YouTube, especially those appearing aimed at children. Examples include clips like Mum cat saves kitten from deadly belly parasites and a short showing a woman eating a parasite, turning into a giant monster, and being healed by Jesus.

YouTube removed the channels for violating community guidelines, saying it is focused on providing high-quality content and reducing the spread of low-quality AI material.

Yet the relentless flow of AI slop continues. Even lifestyle platforms like Pinterest have been affected, prompting the introduction of an opt-out system for AI-generated content, which relies on users admitting their images are AI-made.

Audience reactions and mental impact

Emily Thorson, associate professor at Syracuse University, says reactions depend on how people use social media. Entertainment-focused users may only care whether content is engaging, while those seeking information or community may see AI slop as more problematic.

“If something is clearly a joke, people accept it. But content designed to deceive can anger users,” she explains. Constant exposure to low-quality AI content may reduce attention spans and contribute to “brain rot,” making users quickly consume content that is unlikely to be real or meaningful.

Risks beyond harmless slop

Some AI-generated material can be more harmful. For instance, digital manipulation was recently used to target vulnerable groups online, while fake videos circulated during geopolitical events can mislead the public. Experts argue that platforms need systems allowing authentic content to prove its origin.

Social media companies, however, have reduced moderation teams and increasingly rely on users to flag false or misleading content.

Is a slop-free social media possible?

Creating a platform free from AI slop seems difficult. Machines struggle to detect AI content accurately, and judging what counts as low-quality is subjective. Yet if a new platform emerged and attracted enough users, it might push existing platforms to change, similar to how BeReal encouraged more authentic social media sharing.

For now, Théodore sees AI slop as the new normal. Despite submissions from his followers, he posts less frequently and has mostly accepted the reality of online life.

“I’m not against AI itself,” he says. “I’m against the pollution online of AI slop created purely for quick views and entertainment.”