Back to the archive

The encyclopedia · Advertising & PR · Marketing decision · 2026

Equinox's AI ad showed an Asian woman's head mounted on a stand — it ran for months

Equinox's AI ad showed an Asian woman's head on a stand with a thumb in her mouth. It called the criticism manufactured outrage, then apologized months later.

Equinox · Angry Gods · 2026-01

What happened

In January 2026 Equinox launched 'Question Everything But Yourself,' a global campaign made with agency Angry Gods. One AI-generated image showed a realistic Asian woman's head mounted on a stand, a person's thumb in her mouth, isolated from the campaign's other subjects, who appeared exercising or posing naturally.

Asian American communities on Reddit flagged the image within the campaign's first weeks, arguing it isolated and objectified an Asian woman specifically, echoing a pattern critics call techno-orientalism — depicting Asian people, especially women, as hyper-technologized and dehumanized. The image stayed up at Equinox locations and on social media for months regardless.

When CNN put the criticism to Equinox, a spokesperson initially accused a communications expert who raised it of manufacturing outrage 'for clicks' and imposing a 'salacious interpretation' while 'intentionally ignoring' the work's context. Equinox removed the image more than a week before its eventual statement, and on 29 July 2026 apologized: 'We recognize there was an unintended impact, and we apologize for that.'

Why it happened

  • The campaign's one AI-generated, isolated image happened to depict an Asian woman in a way none of the other subjects were shown, and nobody on the review chain flagged that asymmetry before launch.
  • Equinox's first response to the criticism was to discredit the critic rather than examine the image, delaying any real review by the time it took the dismissal to fail.
  • Letting the image run for months after the first complaints turned a fixable pre-launch miss into a drawn-out story about how the brand handled being told, not just what it published.
What it costimage pulled after months; public apologyembarrassing

The lesson

Attacking the person who flags a piece of creative is a second decision, made in public, and it's usually worse than the first. Review the image again before deciding the critic is the problem.

Aftermath

The episode was widely covered as a case study in how not to respond to a bias complaint — dismiss first, remove quietly, apologize only once the story wouldn't go away — rather than in the AI image itself, which by the time of the apology had already been down for over a week.

Sources

spotted an error? The club wants to know.

Comments · 0

    Sign in to join the comments.

    More like this

    Somewhere, someone solved the problem this company failed at. 2nd Opinion →