AI slop is a pejorative term for low-quality content, including text, images, audio, and video, that is generated in bulk by AI tools with little human curation or editorial effort, then published or shared with the aim of generating engagement, ad revenue, or search visibility rather than genuine value for an audience. The term entered widespread use around 2024, drawing an analogy to "slop" as low-grade, undifferentiated feed poured out indiscriminately, and was applied across platforms including Facebook, Amazon, YouTube, and search results.
Highly visible early examples included surreal AI-generated images on Facebook that nonetheless attracted large volumes of engagement from older or less digitally literate users, sometimes called "Facebook AI slop" or, after one viral example, "Shrimp Jesus" content, as well as a wave of low-effort AI-written e-books flooding Amazon's Kindle store.
Drivers
The falling cost of generating fluent text and plausible images using large language models and text-to-image tools removed the labor cost that had previously limited how much low-quality content any single actor could produce. Platform engagement-ranking algorithms often reward high-volume, attention-grabbing content regardless of quality or originality, creating a direct financial incentive to mass-produce AI content optimized for the feed rather than for readers. Search-engine and AI-assistant visibility incentives, discussed under generative engine optimization, created a parallel incentive to publish large volumes of AI-generated pages aimed at ranking rather than informing.
Platform and industry responses
Several publishers and AI developers introduced labeling and disclosure requirements for AI-generated content, and YouTube updated its Partner Program monetization policies in 2024 to specifically address "inauthentic" mass-produced and repetitive AI content. Search engines adjusted ranking algorithms to demote low-effort, formulaic AI-generated pages, though detection remained imperfect and adversarial, as content producers adapted to evade filters. Some researchers have connected large-scale AI slop to the "model collapse" risk in synthetic data research, in which future models trained on a web increasingly polluted by prior AI output could degrade in quality over successive generations.
Cultural impact
AI slop became a recurring reference point in broader anxieties about the internet's information quality, feeding into and reinforcing popular discussion of the dead internet theory, and prompting renewed interest in curated, human-vouched, or provenance-verified content as a differentiator against an increasingly automated content supply.