Meta has shut down a feature of its new Muse Image model after massive backlash within days. The function allowed users to generate AI photos of Instagram users – without their consent.
The feature worked deceptively simple: users could use other people as templates for AI-generated images by mentioning their public Instagram account (@-mention). No consent required, just a username. Particularly problematic: the feature was enabled by default. Anyone who didn't want their photos used this way had to manually opt out through Instagram settings.
The essentials
- Meta shut down the Muse Image feature that enabled AI portraits of Instagram users without consent
- The feature was enabled by default – opt-out instead of opt-in was required
- Meta admitted: "this feature missed the mark"
- The shutdown occurred days after announcement under pressure from critics
The data protection problem
The concept was problematic from the start. Meta claimed it wanted to offer "a useful creative tool and to give people control over whether their public content could be referenced in this way." But reality looked different: instead of providing genuine control, Meta forced users into a defensive position of having to actively opt out.
In Europe, the feature would not have survived anyway. Stricter data protection rules – particularly GDPR – question such practices. Meta likely anticipated this and pulled the plug before external regulatory pressure mounted.
Parallels to OpenAI – and a cautionary tale
According to reports, Meta may have borrowed the idea from OpenAI. OpenAI's Sora app offered something similar: users could create "cameos" of themselves and – with permission – allow others to use them in videos. The crucial difference: Sora required consent. Yet the feature lost interest quickly once the initial hype faded.
What this means for companies
The swift reversal demonstrates how critical consent-by-design is for AI features – especially when personal data is involved. Companies working with AI should take this as a signal: opt-in instead of opt-out is not only ethically sound but increasingly a regulatory expectation. The EU AI Act and GDPR make clear that data protection is not an obstacle to innovation but its foundation. Ignoring this risks not only reputational damage like Meta faced, but also significant penalties.
Sources
Editorially owned by Ideal Syka. Sources and method: Newsroom & method. Tips and corrections: ai@i6eal.de.
