The motivation is almost always money
In the evolving landscape of digital identity, an Australian content creator found her livelihood suspended not for her own actions, but for those of an AI-generated impostor wearing her face — a face Meta had verified and rewarded with half a million followers. The incident, resolved only through journalistic intervention rather than platform accountability, reveals a deepening fracture between the sophistication of AI-driven deception and the institutional will to confront it. As the tools of impersonation grow cheaper and more convincing, the question is no longer whether platforms can moderate at scale, but whether they have chosen to.
- An AI-generated doppelgänger, verified by Meta and followed by 500,000 people, was selling subscription content under a real creator's face while the real creator's account was banned for the impostor's violations.
- Every avenue of recourse — appeals, reports, follower campaigns, automated support — looped back to silence, leaving the catfish online and profitable for over a month.
- Only a journalist's phone call to Meta unlocked what hours of navigation through AI chatbots and automated emails could not: a human decision and a restored account.
- Meta cites the removal of 20 million impostor accounts in 2025 and a 33% drop in reports, but critics argue the system is built to handle volume, not to protect the individual creator who falls through the cracks.
- Australia's Telecommunications Industry Ombudsman, armed with data showing 80% of Australians faced digital platform problems last year, is pushing for regulatory authority over social media — authority it does not yet hold.
- Experts warn the window for platform self-correction is narrowing as AI impersonation tools accelerate faster than the moderation systems designed to catch them.
Alicia had built a career across YouTube, Twitch, and subscription platforms making ASMR content. Impersonation was nothing new to her — but what she discovered in May was different in kind. Her Instagram had been banned for violating community standards. In its place, a Facebook account using AI-generated images of her face had accumulated more than 500,000 followers. Meta had verified it. The impostor was collecting subscription payments from people who believed they were paying her.
What followed was a bureaucratic maze. Appeals went nowhere. Reports of the fake account produced no action. An AI chatbot handled her initial requests. When she eventually reached a human representative — after creating a new account, getting it verified, and spending hours in automated systems — she received a promise of investigation and then a month of form emails. The catfish remained online, still earning, still wearing her face.
Alicia was, by her own acknowledgment, fortunate. She had the financial cushion to absorb the disruption and the resources to keep pushing. Most creators do not. For them, a suspended account is not an inconvenience but a severed income. Her account and the impostor's were ultimately dealt with only after a journalist contacted Meta directly — a phone call that accomplished what the platform's own systems could not.
Meta points to the removal of more than 20 million impostor accounts in 2025 and a reported 33% decline in impersonation complaints. But Alicia's case suggests the system responds to external pressure rather than internal accountability. New AI support tools are being rolled out, including one launched in Australia in March, but these are designed for volume — to deflect, not to resolve.
The broader picture is stark. Research from Australia's Telecommunications Industry Ombudsman found that 80% of Australians encountered a digital platform problem in the past year, more than half involving account access. Two in five people never complained at all — they simply gave up. Of those who did, two in three left unsatisfied. The ombudsman currently holds no authority over social media companies, but Cynthia Gebert is pushing to change that, arguing that resignation is no longer an acceptable response when so much of modern life depends on platform access. Whether the platforms will act before regulation compels them remains the open question.
Alicia had built her living on the internet. She made ASMR content across YouTube, Twitch, and subscription platforms, cultivating audiences who paid for her work. She was also used to being stolen from. Impersonators had lifted her images before. But in May, when her Instagram account vanished—banned for violating Meta's community standards—she discovered something far more sophisticated waiting in its place.
Someone had created a Facebook profile using her face, except the face wasn't quite hers. The images were AI-generated, stitched together from multiple prompts, her features blended with other bodies, other people. The fake account had accumulated more than 500,000 followers. Meta had even verified it. The impostor was selling subscription content to people who thought they were paying Alicia.
When she tried to reclaim her identity, the platforms offered her a maze. She appealed her ban. She reported the fake account. She asked her followers on X to flag it. Nothing worked. The catfish stayed online, still collecting money, still wearing her face. "The motivation is almost always money," she said, stating the obvious truth that the platforms seemed unable or unwilling to act on.
Meta's support system sent her in circles. An AI chatbot fielded her initial requests. When she finally reached a human being—only after creating a new account, getting it verified, and spending hours navigating automated systems—that person promised to investigate. Then came a month of automated emails. No resolution. No action. No human on the other end.
Alicia was fortunate. She had enough financial cushion to absorb the lost income, the disrupted business, the stolen audience. She could wait out the bureaucracy. She could hire someone to help navigate it. Most creators cannot. For them, a banned account isn't an inconvenience—it's a severed livelihood. Meta eventually restored her account and removed the impostor only after a journalist from triple j hack contacted the company directly. The platform's own systems had failed entirely.
Meta claims it removed more than 20 million accounts impersonating large creators in 2025. It says reports of impostor accounts are down 33 percent. It uses a mix of automated and human review, the company says, and it encourages users to report violations. But Alicia's experience suggests the system works only when external pressure is applied—when a news organization makes a phone call that an individual creator cannot.
The vulnerability is widening. As AI tools become more sophisticated, more accessible, more convincing, creators across the platforms are warning that catfishing will accelerate. The technology is outpacing the moderation. The platforms know this. They are rolling out new AI support assistants—in Australia's case, one launched in March—but these systems are designed to handle volume, not complexity. They are built to deflect, not to solve.
Meanwhile, research from Australia's Telecommunications Industry Ombudsman reveals a broader crisis in digital accountability. Eighty percent of Australians experienced a problem with a digital platform in the past year. More than half involved account access. Two in five people never even filed a complaint—they simply gave up. Of those who did complain, two in three walked away unsatisfied. The ombudsman currently has no power over social media platforms, only over telecommunications companies. But Cynthia Gebert, the ombudsman, is pushing to change that. "I don't know that that's good enough that we do just give up when so much of our lives are online now," she said. The question is whether the platforms will wait for regulation to force their hand, or whether they will act before the technology they've enabled becomes impossible to control.
Notable Quotes
Every photo, I can tell it looks like me, but not quite. It looks like maybe they're using multiple prompts, so like maybe someone else's body with my face.— Alicia, content creator
I don't know that that's good enough that we do just give up when so much of our lives are online now.— Cynthia Gebert, Telecommunications Industry Ombudsman
The Hearth Conversation Another angle on the story
Why did Meta's own systems fail to catch this? They have the resources, the data, the technology.
Because catching it requires human judgment at scale, and humans are expensive. The AI chatbots are cheap. They're designed to handle volume, not to actually solve problems. When Alicia's case required real investigation, the system had no incentive to provide it.
But Meta verified the fake account. That's not a failure of detection—that's a failure of verification itself.
Exactly. Verification became a rubber stamp. It signals legitimacy to users, which makes the fraud more effective. The verification process probably relied on the same automated systems that missed the impersonation in the first place.
What would actually fix this?
Humans. Real people reviewing real cases. But that costs money. It's cheaper to remove accounts after journalists call, to issue statements about how many accounts were removed, to roll out new AI assistants that sound like progress.
Is this specific to Meta, or is it an industry problem?
It's an industry problem. But Meta is the one with 500,000-follower fake accounts still collecting money. That's not a small failure.
What about the creators who don't have a journalist's phone number?
They give up. They lose their income. They move on. That's the system working as designed—just not for them.