AI thirst – A shirtless TikTok performer with rapidly growing followings may not exist at all. Digital forensics traced the account style to AI-generated deepfakes, while adult entertainers warn that copycat imposters are being used to lure fans into payments—often throug
On a scroll that’s measured in seconds, Derek Lam—shirtless, dancing, and never speaking—built an audience fast. He has more than 31,000 followers on TikTok and nearly 40,000 on X as of this writing. His videos are brief, lasting just seconds. His X selfies. viewers say. don’t match the man fans think they’re watching: they show a different person from about three years ago.
In the comments, admirers respond as if he’s real and as if he’s the first of his kind. Compliments arrive in multiple languages. including “hombre bello y sensual. ” alongside superlatives like “this might be the finest man on the internet. ” with emoji ranging from red hearts to crying-laughing faces and lips. The engagement has helped create a market for what he presents as “exclusive,” seemingly adult content.
But digital forensic analysis described in a new report suggests the videos may be AI-generated, part of a rising class of “deepfakes”—media that resembles real people, made by artificial intelligence.
A key detail is silence. Derek never speaks. Another is repetition that doesn’t behave like a living body. While many real hot creators might convert follower growth into brand deals or “get ready with me” content. Derek’s output stays in the same narrow channel: short clips. the same style of flexing and movement. and an almost too-perfect aesthetic.
That sameness became clearer when the report examined accounts he follows. One account. which goes by the name Vance Ford. appears to have tens of thousands of followers and nearly identical videos to Derek. The flexing. dances. movements. and the music they’re set to were described as the same. but performed by what appears to be a different man.
The forensic trail pointed back to a likely shared method. Siwei Lyu. a professor at the University at Buffalo who studies digital forensics. reviewed Derek’s videos frame by frame and flagged multiple “AI tells.” Those included a distorted watchface with swirls instead of numbers. and a moment where the fingers on one hand were reportedly all the same length. Lyu also pointed to chest hair that fluctuates—dense in one frame and dissipating in another.
Lyu also analyzed the Vance videos with manual review and AI-detection software. He confirmed that “their movements are nearly identical — consistent with generation from a shared motion source,” and noted moments of distortion, unintelligible text, and facial warping.
Lyu is also the scholar the report describes as explaining how the technology improved from early deepfakes that often required clunky apps and large amounts of data to generate low-quality results. When deepfakes first appeared in late 2017. the report says. they were often easy to spot because celebrities or politicians were pasted onto bodies with obvious artifacts. Lyu described those first set of deepfakes as being used to create pornographic videos that replaced subjects with the faces of celebrities.
But Lyu said the newer methods changed the playing field. The report describes him saying the latest systems do not need as much data to train a model. “Some of the most recent algorithms just need a single picture — just a single picture of someone,” he said. He also described AI programs that can change a person’s appearance and voice in real time. like in Facetimes and Zooms or during live broadcasts.
That shift matters because it makes a convincing social media presence cheaper and easier to generate. With people posting photos and videos constantly, the report says it’s now “extremely easy” to create a convincing profile for someone who doesn’t exist, then use it to catfish unwitting people.
Lyu’s warning was blunt: “It’s becoming more and more challenging to visually tell deepfakes apart.” He added that when he started working in this area seven years ago, checking was not as difficult.
Efforts to contact both Derek and Vance were unsuccessful within the report. After social media messages were sent to the owner of Derek Lam’s account with evidence described from Lyu. there was no response. The report says it also attempted to contact Vance through direct messages and received no response. and that two models who appear to be the real people Derek and Vance were trained on did not respond.
For fans, the discovery can land with an abrupt emotional whiplash: humiliation, anger, and a sense of being made small.
Patrick. one of Derek’s followers on X. told the reporter that he had no idea the account might be a deepfake. After learning it was likely artificial. Patrick said he felt older and more vulnerable—vulnerable not just to scams. but to “light financial crime.” He is 33. about 30 years younger than the “boomer” he joked about. but the report describes how being fooled by a hot AI man made him feel old and exposed.
“It’s embarrassing,” Patrick said, framing his reaction around the fact that he hadn’t known he was following a fake. He said he uses AI in everyday life only for organizing and writing emails. As for Derek’s kind of content, he said the possibility of seeing more of it doesn’t thrill him.
Others described a similar mixture of discomfort and skepticism, even when they admitted being lured in by the fantasy.
Chris. also 33. said he had seen other AI accounts and that Derek was “really good.” But he said the issue wasn’t that he follows attractive men online. The issue was how he felt after realizing he’d been duped: “You can see there’s like no life in his eyes.” Chris works in digital marketing. He told the reporter he has seen AI used professionally to tabulate calculations for campaigns. and that he has used it privately for memes. What he finds scary is that “everybody has access to it.”.
Chris also suspected a more deliberate scheme. He argued that whoever is running Derek might have hijacked the username—meaning the original person Chris says he followed—and then replaced it with AI-generated content to drive up follower counts. Chris said he has seen scams like that online. and described a scenario where users could end up chatting with the person behind the deepfakes and then be pulled into subscribing to fake content and giving personal information. including photos or perhaps passwords.
“If someone’s making an AI of me doing double anal,” Chris warned in the extreme version of the danger he described—“I should be making the money,” that report later quotes Cherie DeVille as saying, underscoring how real performers view these fakes as both theft and a business threat.
DeVille, an adult star who calls herself “The Internet’s Stepmom,” has roughly 4.5 million followers on Instagram. The report describes her account as often down. and she says that is linked to fraudsters who impersonate her in order to send traffic to AI imposters—aimed at getting her real account removed.
“They want to be the biggest me. They want to be the biggest scammer,” DeVille said. She described a pattern: fake accounts of her report her, and fraudsters use altered AI images to scam fans without her real account getting in the way.
DeVille and others in the report say the playbook is familiar. It involves fake photos or videos of DeVille—or another star—followed by an impostor profile created to trick fans. often through social media direct messages. Later, fans are squeezed for money through payment methods like PayPal or Amazon gift cards, with offers for unique content.
“If you made a fake me and I don’t do double anal, but my AI can, they could have all kinds of ‘exclusive’ stuff,” DeVille said, describing the kind of “exclusive” content that could be used as bait.
For performers, the legal and practical questions are tangled. The report notes that not every deepfake is a carbon copy. Some creators may use a face from one actress, a torso from another, and legs from a different star. That makes fakes harder to track down and harder to fight legally.
Rachel Steele, an adult star and CEO of Red MILF Productions, asked questions that go beyond a single scam. In an email, Steele asked: “Who owns your face once it’s scraped into AI systems?. Who profits from your digital clone?. How do performers protect themselves from unauthorized replicas or manipulated content?” Steele said those questions are still “very unanswered.”.
Rachel Steele’s concerns echo those voiced by Raissa Bellini. an OnlyFans creator who touts gymnastics and firebreathing among her skills. Bellini told the reporter that real creators face an impossible comparison—characters that can be flawless in every image. never age. never have bad lighting. never get tired. and can appear available 24/7. She described seeing people create AI personas based on popular models or influencers by tweaking small details like hair color or eye color.
OnlyFans said it has rules meant to prevent this kind of impersonation. A spokesperson told the reporter via email that the company’s terms of service prohibit deceptive or inappropriate content. The spokesperson also said that content posted on OnlyFans must belong to a verified 18+ OnlyFans content creator: “This means that you can only share content which has been generated. altered or enhanced by AI if it clearly features the verified OnlyFans creator and the user can tell that the content has been generated. altered or enhanced by AI.”.
Bellini said that while OnlyFans has measures to protect creators, smaller subscription and adult-content platforms may not have the same guardrails. She also said she has seen social media platforms’ algorithms appear to favor AI over human creators.
For legal experts, the barrier is that existing laws weren’t written with AI clones in mind. Jason Schultz. a law professor and director of NYU’s Technology Law & Policy Clinic. told the reporter that copyright and right of publicity laws have generally protected humans over the last couple of centuries. But he said courts now have to interpret those laws in light of new technology and other existing rights. including free speech.
Schultz said there are more than 100 current cases pending about training AI with copyrighted material. He also pointed to the difficulty of determining whether an AI-generated persona violates someone’s right of publicity. especially when the human involved isn’t a celebrity. He described it as “more clear-cut when the human involved is a celebrity” because their appearance and public persona are distinct. while it becomes murkier when the AI creates an amalgam rather than a one-to-one copy.
He explained that could raise questions about whether avatars are based on a particular entertainer or whether they are aggregates.
Even if courts side with humans, Schultz cautioned that technology will continue moving faster than court decisions. “I think that the thing that worries me a little is we’re going to get these sets of decisions in two years. but we’ll be dealing with the next three generations of technologies. ” he said.
For DeVille, the stakes aren’t theoretical. She told the reporter that if her income started tanking and the theft reached a point where she couldn’t compete with herself. she might have no choice but to retire. She said she isn’t against AI in principle—she wants to be in control of it. owning her likeness. voice. and image. and choosing what happens with it. She also said she wants compensation or legal protection if someone uses Cherie DeVille without permission.
“It would be a beautiful way to extend my career beyond what my knees can take,” DeVille said.
But if someone is making an AI version of her doing double anal, she added, “I should be making the money.”
For fans. the warning is simpler: if you can’t tell whether the person on the screen is real. you can’t tell what you’re being asked to pay—and what your information could become. The report described multiple red flags offered by deepfake experts and adult entertainers: checking logos or objects with text like clocks and posters for distorted. jumbled numerals; watching for inconsistent backgrounds with unusual blur; assessing whether the account is associated with OnlyFans. where adult entertainers said there are rules on AI and an ID verification process; and treating off-platform payment requests—like Amazon gift cards—as a major suspicion signal.
One adult entertainer told the reporter, “I don’t need an Amazon gift card,” pointing out the off-platform nature of such requests. Other red flags included requests for private information such as bank details or passwords.
Patrick captured the mood after he realized the truth. He said a person being real—someone you could run into at a bar—is “half the fun. ” and that AI porn was not of interest to him. But he also described the emptiness of being lured in by something that never existed. “Not being able to tell the difference. ” he said. left him feeling old and “hollow. ” a reflection of unattainable desire that machines can now mimic and deliver undetected.
On an internet built for speed and certainty, the Derek Lam story lands as a reminder: the modern scam doesn’t always require a con artist with a voice. Sometimes it’s just a body, a beat, and a silence that never answers back.
AI deepfakes TikTok X adult entertainment OnlyFans right of publicity digital forensics scams Amazon gift cards