AI slop, sometimes called digital slop, describes AI-generated content that isn’t technically broken, the grammar is correct, the image renders cleanly, but reads as generic, derivative, and evidently produced without anyone’s judgment applied to it. It shows up in marketing feeds, search results, and social platforms wherever volume got prioritised over a reviewed, specific final output.
The objection audiences raise to it isn’t really about the technology. Survey research on AI-generated advertising consistently finds people can’t reliably identify AI-made content on sight, yet they can tell, from tone and specificity, when nothing about a piece was decided by a person who cared how it turned out. That distinction is what erodes trust over time, not any particular technical flaw, and it compounds: platforms and search engines are increasingly penalising exactly this kind of low-effort, high-volume output as it floods their feeds and results.
The fix isn’t avoiding AI in production. It’s keeping a visible, specific human decision at the point the audience sees the work, the final image chosen, the claim a piece of writing makes, even when AI did the drafting, resizing, or variation-generation underneath it.