Meta's AI image tool for Instagram lasted just days before a privacy revolt killed it
By isabelle // 2026-07-13
Copy
 
  • Meta launched an AI image tool using public Instagram accounts without user consent.
  • The feature was automatically enabled for public accounts, requiring manual opt-out.
  • SAG-AFTRA and privacy advocates condemned the tool as exploitative and dangerous.
  • Critics warned the feature enabled non-consensual image misuse, fraud, and scams.
  • Meta pulled the feature after a few days following widespread backlash and criticism.
Meta, the tech giant that owns Facebook and Instagram, launched a new artificial intelligence feature on Tuesday that let users generate images by referencing public Instagram accounts. By Friday, the company had pulled the plug. The rapid reversal came after a wave of criticism from privacy advocates, a Hollywood union, and individual users who said the tool violated basic expectations of consent related to personal images. Meta introduced Muse Image on Tuesday as the first image-generation model from Meta Superintelligence Labs. The feature was integrated into the Meta AI chatbot and allowed users to @-mention a public Instagram account to generate new images based on that person's photos. The company also rolled out more than 30 new AI-powered effects for Instagram Stories as part of the same launch. The problem, critics said, was that the feature was automatically enabled for users with public accounts. Users had to manually opt out if they did not want their images used as raw material for AI-generated content.

Hollywood union and privacy advocates push back

SAG-AFTRA, the union representing actors and media professionals, quickly urged members and other Instagram users to opt out. "Anything other than a clear and conspicuous opt-in for these types of uses of Instagram users' images is unacceptable, and an utter miscalculation of public sentiment regarding the obvious dangers and harms inherent in such use," the union said. Emmy-winning actor Hannah Einbinder, known for the series "Hacks", publicly criticized the feature on Instagram, noting it had been turned on automatically and urging users to disable it. Privacy International called the tool "the latest sign AI companies see people's images and data as raw material to be exploited." Donald Campbell, advocacy director at the tech justice nonprofit Foxglove, called the rollout an "obvious recipe of disaster," pointing to what he described as a catalogue of harms from nonconsensual AI-altered images on social platforms over the past year. "It is hard to see why Mark Zuckerberg thinks facilitating yet more of this creepy image manipulation is a good idea," he said.

Researchers and advocates warn of fraud and exploitation

Neal K. Shah, an NIH-funded caregiving researcher who runs the caregiving company CareYaya, told Fox Business that scammers have already hijacked his face. AI-generated advertisements circulated showing him endorsing supplements he had nothing to do with — products marketed with false claims about treating dementia. His own followers started reaching out to ask whether the endorsements were real. "All of these older people have been scammed, and my image has been used to scam them, and I can't do anything about it," Shah said. He has taken to warning viewers in his own videos not to trust ads that appear to show him selling anything, and says he reported the fakes to Meta repeatedly without result. Shah's experience is the whole argument in miniature. The people most easily fooled by a fabricated endorsement are the elderly — and they were never asked whether they wanted to live in a world where a stranger's face could be borrowed to sell them a pill.

Meta's defense does not survive contact with its own reversal

Before the retreat, a Meta spokesperson insisted the company had "built Muse Image with strong controls and safety guardrails from day one," noting that private accounts and accounts belonging to minors were excluded and that adults could opt out "with just a couple clicks." But guardrails built from day one do not get dismantled on day four. In a statement, Meta said its intent was to provide a useful creative tool and to give people control over whether their public content could be referenced in this way. The company acknowledged hearing feedback that the feature "missed the mark" and confirmed it is no longer available. SAG-AFTRA welcomed the reversal. "With the dangers of nonconsensual digital replicas well known to all, a feature that encouraged that behavior is unwise. We appreciate its discontinuance. It is the responsible thing to do," a union spokesperson said. Meta's remaining AI creative tools are untouched, and the company has lost nothing but a few days of favorable coverage. What should trouble Americans is the default. Consent was not asked for; it was assumed, and buried in a settings menu most users would never find. A company only surrenders that kind of advantage when enough people notice — which means the burden fell, as it usually does, on ordinary users to defend a right they should never have had to claw back. Sources for this article include: Reuters.com TheTab.com TechCrunch.com FoxBusiness.com
Copy