Document Type
Working Paper
Publication Date
9-2026
Language
en-US
Abstract
The term “AI slop” has become popular to describe the output of generative AI systems seen as voluminous, low quality, or the result of little effort. When AI-generated music and videos flood platforms, they are called slop. Peer-reviewed journals and legal tribunals are drowning in low-quality and low-reliability AI slop submissions. Employees are seen to be producing mountains of slop in their reports and communications with each other. The term has inertia and heft, and the phenomenon has significant consequences. Most of them are not good.
But the boundaries of “AI slop” and its usefulness in policy discussions are not clear. What distinguishes slop from other machine-aided human authored work? Does the concept of “AI slop” help policymakers and researchers better understand a problem or identify a solution? This Article interrogates the concept of AI slop to identify its essential features and document how it operates in three cultural domains where it is commonly deployed—creativity, research, and work.
We propose a provisional definition of slop as any output of a probabilistic automated system produced with little exertion that asymmetrically burdens recipients and tends to degrade cultrual domains. We also conceive of AI slop along a spectrum. Works can be more or less “sloppy,” depending on the relative presence of its three constituent parts: negligible exertion, asymmetrical imposition, and domain degradation. We conclude that AI slop is a useful and important frame in AI discourse and policymaking because it provides a shorthand to distinguish harmful, inferior AI output from meaningful human authorship and because it highlights the pathologies of generative AI for social institutions.
Recommended Citation
Jessica Silbey & Woodrow Hartzog,
AI Slop
(2026).
Available at:
https://scholarship.law.bu.edu/faculty_scholarship/4308
