Gizmodo's Webb Wright Argues AI Slop Is a Two-Way Mirror in 'Society of the Psyop'

What happened

On June 28, 2026, Gizmodo published a 7-minute essay by Webb Wright titled "When You Gaze Into the AI Slop, the AI Slop Also Gazes Into You," with the dek "Welcome to the 'Society of the Psyop.'" The piece is an opinion and analysis article rather than a breaking-news report, and it draws on art, philosophy, and the history of surveillance technology to make its argument.

Wright opens with a question — "What happens when you look at an apple?" — and contrasts a materialist explanation (photons, photoreceptors, a visual representation of a real object) with a more philosophical or artistic reading, in which the apple becomes a portal into personal memory and association. This framing sets up the essay's broader thesis: that looking at AI-generated content is not a passive act of consumption, but an encounter that also shapes the viewer.

The article references Trevor Paglen's 2020 artwork "They Took the Faces from the Accused and the Dead," which documents how modern facial-recognition research in the 1990s relied on image databases built from mugshots and prison records supplied by the U.S. National Institute of Standards and Technology (NIST). Paglen's work is used to illustrate how the datasets underlying AI systems carry the politics of their origins — a point Wright extends to the current era of generative AI.

The full body of the essay was not available in the captured source material beyond the opening sections, so the complete arc of Wright's argument — including specific examples of AI slop, named platforms, or policy recommendations — could not be independently verified from the supplied inputs.

Why it matters

The phrase "AI slop" has become shorthand for the flood of low-effort, AI-generated text and images filling social feeds, search results, and comment sections. Wright's essay pushes beyond the common framing of slop as a quality or spam problem and argues that it functions as a psychological operation — a "psyop" — that works on the viewer even as the viewer scrolls past it.

By connecting contemporary generative AI to the history of facial-recognition datasets built from incarcerated people, the essay situates current AI content within a longer lineage of surveillance and control. The implication is that the infrastructure behind AI slop — the training data, the models, the recommendation systems that distribute it — is not neutral, and neither is its effect on the people who encounter it.

For anyone building, distributing, or regulating AI systems, the essay raises a question that is easy to dismiss as philosophical but has practical consequences: if AI-generated content changes the viewer as much as the viewer consumes it, what design responsibilities follow?

What to watch

• Whether Wright's "Society of the Psyop" framing gains traction as a term for describing the psychological impact of AI-generated content at scale.
• How regulators and platform designers respond to arguments that AI slop is not merely a quality issue but a structural one tied to training-data provenance.
• Whether further reporting connects specific generative-AI products to the surveillance-dataset lineage Paglen's work documents.
• The full text of the essay, which was only partially available in the captured sources, may contain additional claims or examples worth tracking.