Slop, consent, and the fight over the human

The same period produced the backlash, and it sharpened into two fights: one over quality, one over consent.

The quality fight gave the era its word. 'SlopSlopAI content shared without human review or intention, defined less by the tool that made it than by the missing step of someone curating it before it goes out.,' borrowed from an old term for low-grade filler, was popularized by the developer Simon Willison in 2024 with a simple definition: slop is AI content shared without review, and 'sharing unreviewed AI-generated content with other people is rude.' By late 2025 dictionaries were naming slop a word of the year. The scene's defensive claim, after the Sora app flooded feeds with disposable clips, is that slop is a choice, not a property of the medium: volume, zero intention, and no human review are what make slop, not the tools.

The consent fight is sharper because it is about people. Artists built defenses, the GlazeGlazeA tool from the University of Chicago that cloaks an artist's images so models cannot easily learn and copy their style. and NightshadeNightshadeA tool from the University of Chicago that subtly poisons images so that models training on them without permission are degraded. tools from the University of Chicago that cloak or poison images against training. Performers drew lines. When a synthetic 'actress' named Tilly Norwood was floated to talent agencies in 2025, SAG-AFTRASAG-AFTRAThe United States labor union representing actors and other media performers, whose agreements require consent and compensation for any digital replica. answered flatly that she 'is not an actor,' but 'a character generated by a computer program that was trained on the work of countless professional performers without permission or compensation.' When OpenAI's Sky voice sounded like Scarlett Johansson, who had declined to license it, she said she was 'shocked, angered and in disbelief.' When Sora 2 launched with an opt-out policy, it filled with deepfakes of dead celebrities until the estates and the union forced a reversal.

Around the fight, an infrastructure of consent grew. Talent agencies built likeness vaults, CAA with Veritone and WME with Vermillio. ElevenLabs licensed the voices of dead stars from their estates. SAG-AFTRA signed sanctioned deals for consented voice models. The market's answer to the question 'did a human agree to this' is becoming a product category of its own.

The hardest question, jobs, has the least settled answer. An industry-commissioned 2024 study projected as many as two hundred thousand United States entertainment jobs disrupted within three years, and the figure is quoted constantly, but as of mid-2026 no rigorous accounting has confirmed it, and the one large visual-effects collapse of the period, Technicolor's, was driven as much by debt and the post-strike slowdown as by AI. What is confirmed is smaller and more concrete: Netflix using AI on three hundred titles, advertising budgets falling from millions to thousands, one-person studios doing what once took a crew. The disruption is real; its total size is still being written.