AI-generated influencer accounts on Instagram are no longer a small curiosity. According to the source investigation, hundreds of these accounts are using images, videos, and deepfake techniques to build audiences, promote paid content, and compete with the real models and adult content creators whose work is often being reused without permission.
The practice was first reported by 404 Media in April and has since grown into what the investigation describes as an industrialized content system. The accounts are built with off-the-shelf AI tools and apps, some of which are available through the Apple App and Google Play Stores.
How the AI influencer accounts work
The basic model is simple: create an AI-generated persona, post images and videos that appear to show that persona, then direct followers to monetized links. Those links can lead to dating sites, Patreon, OnlyFans competitors, and AI apps.
In many cases, the content is not entirely synthetic. The investigation found accounts that take videos from real models and adult entertainment performers, then replace the original person’s face with an AI-generated face. The result is presented as new material from a virtual influencer, even when the body and original performance came from a real person.
That distinction matters because it changes who benefits from the attention. A human creator may have made the original video, but the AI account can capture the audience, the follower growth, and the paid clicks. For creators who already rely on Instagram visibility to reach subscribers or fans, that creates direct competition from accounts that can be produced quickly and at scale.
What the review found
The investigation reviewed more than 1,000 AI-generated Instagram influencer accounts. Out of those, 100 included at least some deepfake content, usually by taking existing videos and replacing the face with an AI-generated one. The remaining 900 accounts shared images that were entirely AI-generated, though some were trained on real photographs and some were made to resemble celebrities.
The review also found a disclosure problem. Of the 100 accounts that shared deepfake or face-swapped videos, 60 identified themselves as AI-generated in their bios, using descriptions such as “virtual model & influencer” or “all photos crafted with AI and apps.” The other 40 did not include any disclaimer that they were AI-generated.
One account named “Chloe Johnson” had a verified Instagram account and 171,000 followers before Meta deleted it. A 404 Media reader used Google Lens to locate original source videos for nine of the account’s posts. The investigation said those videos showed the operator swapping the AI influencer’s face onto the bodies of real women, including Tana Rain, Skyler Simpson, and Kyla Yesenosky. Other videos came from TikTok and Instagram users with smaller followings, including Ulia Nova and Annabella Sinclair.
The investigation also found face-swapping accounts sourcing videos from swimwear runway shows and from Getty’s stock image and video site iStock.
The money behind the personas
The business model is not limited to gaining followers. The source article describes AI influencer accounts sending users to paid platforms where fans can buy access or individual pieces of content.
The “Chloe Johnson” Instagram account linked to Fanvue, a site similar to OnlyFans, and also linked to another website where nude photos and hardcore porn videos could be bought individually for between $3 and $22. The investigation said many other AI influencer accounts it reviewed also monetized content on Fanvue.
There is also a market for instructions on how to build these accounts. 404 Media purchased two guides: a PDF instruction manual called Instagram Mastery by an AI influencer agency called Digital Divas, and AI Influencer Accelerator by someone calling themselves Professor EP.
Professor EP says they operate the Emily Pellegrini AI influencer Instagram account, which has 253,000 followers. Professor EP was also a judge in the first “Miss AI” contest, which was made in partnership with Fanvue. The Daily Mail called Pellegrini the “World’s Hottest Model” in January.
After that, the operator of the Emily Pellegrini account shifted from posting as Emily Pellegrini to posting as Professor EP, promoting instruction manuals for what they described as “AI Pimping.” Professor EP claims to have made more than a million dollars in six months. When operating the Emily Pellegrini account and serving as a Miss AI judge in July, he claimed to have made $100,000 on Fanvue alone.
A Fanvue representative told 404 Media, “Yes, Emily had earned this revenue on Fanvue.” Fanvue also said it had no affiliation with the course posted on Instagram or with Emily’s team’s marketing decisions, and said usage of the active Emily Pellegrini Fanvue account is extremely low.
Why human creators are worried
For adult content creators and Instagram models, the rise of AI influencer accounts is not an abstract platform trend. Elaina St James, an adult content creator who promotes her work on Instagram, said she and others are now competing with AI rip-off accounts, many of which use stolen photographs and videos.
St James said other changes to Instagram’s algorithm may also be involved, but that her reach has fallen sharply since the explosion of AI-generated influencer accounts. She said her typical monthly views used to be 1 million to 5 million, but in the last 10 months she has not cracked a million and sometimes comes in under 500,000 views.
“This is probably one of the reasons my views are going down,” St James told us in an interview. “It's because I'm competing with something that's unnatural.”
The concern is not only that AI content exists. The issue is that some accounts appear to use the labor, bodies, and videos of real people while presenting the result as a different, synthetic personality. That can turn a creator’s original material into a competing product.
A preview of a more synthetic internet
Alexios Mantzarlis, the director of the security, trust, and safety initiative at Cornell Tech and formerly principal of trust and safety intelligence at Google, compiled a list of around 900 accounts that 404 Media reviewed. He began looking into the accounts after coming across one while using Instagram casually.
Mantzarlis said the accounts point toward a broader change in social media and the internet, which he described as “a rising blended unreality.” He also said he could have found 900 more accounts, and that the only reason he did not was that Instagram restricted the account he used to scrape the platform.
“It felt like a possible sign of what social media is going to look like in five years,” Mantzarlis said in an interview. “Because this may be coming to other parts of the internet, not just the attractive-people niche on Instagram. This is probably a sign that it's going to be pretty bad.”
The investigation presents Instagram as a testing ground for a larger problem: when AI-generated content is easy to make, easy to monetize, and hard for users to verify, platforms may fill with synthetic accounts that compete directly with human creators. The result is a social feed where authenticity, ownership, and disclosure become harder to read at a glance.