Meta has shut down a feature in its Muse Image model that generated AI photos of Instagram users without their explicit consent, saying the feature "missed the mark." Users could create images by simply mentioning other people's public accounts, with opt-out as the default protection—a design that likely would have violated European data protection rules. The rapid discontinuation underscores the regulatory and reputational challenges of opt-out-based AI systems that use public content.
Summaries like this, in your inbox every morning.
Sign up free →What happened
Meta removed a feature from its Muse Image model that allowed users to generate AI images of other people by mentioning their Instagram accounts. The feature was enabled by default and required people to manually opt out through Instagram settings if they did not want their photos used.
Why it matters
The company acknowledged the feature "missed the mark" and shut it down within days of launch. In Europe, stricter data protection rules would likely have blocked it anyway. For Meta, the rapid reversal signals the reputational risk of opt-out-by-default AI systems that use people's public content without explicit permission.
What to watch
Meta said it had intended to offer "a useful creative tool and to give people control over whether their public content could be referenced in this way" — a goal the opt-out design undermined. The company appears to have borrowed the concept from OpenAI's discontinued Sora app, which let users create "cameos" but required consent before others could use them.
Meta's rapid retreat from the Muse Image feature reveals the tension between offering "useful creative tools" and securing genuine user consent. The company framed its intent as giving people control over whether their content could be referenced, yet the opt-out default—where users had to manually disable the feature through Instagram settings—placed the burden on individuals to protect themselves rather than requiring affirmative permission upfront. This friction between stated intent and actual design is what Meta now acknowledges as the core problem.
The comparison to OpenAI's Sora cameos is instructive. Sora's approach required explicit consent before others could use a person's likeness, and while the feature initially went viral, interest waned over time. Meta appears to have gambled that an opt-out model would drive higher engagement, but public and regulatory backlash—particularly the prospect of European data protection enforcement—made that trade-off untenable. The swift shutdown suggests Meta prioritized avoiding regulatory exposure and brand damage over keeping the feature live while refining consent mechanisms.
No discussion yet for this article
Get curated AI news from 200+ sources delivered daily to your inbox. Free to use.
Get Started FreeFree · takes 30 seconds · unsubscribe anytime
1 minute a day. The AI essentials.
200+ sources · Email / LINE / Slack
