We are excited to announce that PersonaShield is live. For the first time, you can protect your likeness and give fans an official way to create with you, in one place. Here's how it works. 🧵

NIL rights gave athletes the ability to profit from their name, image, and likeness for the first time. That was 2021. Four years later, AI is exposing the parts of that framework that were never built for what is possible now. That reality is playing out in two directions. On one side, athletes can now be cloned into content they never approved, deepfaked into game footage or interviews that did not happen, or tied to positions and products they would never associate with themselves. On the other side, AI can be a real engagement tool if handled right. Officially licensed digital versions for fan interactions, custom content for sponsors, and creative extensions of their brand across markets and languages they could not otherwise reach. What ties both sides together is choice. Whether AI becomes a threat or an opportunity for an athlete depends entirely on who gets to decide how their likeness is used. #NIL #AILikeness #LikenessProtection #CreatorEconomy #Sports personashield.com/blog/ai-de…

Two months. That is how long a Jon Stewart deepfake about Italian Prime Minister Giorgia Meloni has been up on TikTok. No AI label. No takedown. Still running. TikTok's policy requires realistic AI content to be labeled. This one does not carry one, and it is not the exception. Two months is a long time for a deepfake to sit on a major platform. It is more than enough time for the video to reach an audience, for viewers to form impressions, and for the harm to spread past the point where a takedown makes much of a dent. That is the window PersonaShield exists to close. #AILikeness #DeepfakeDetection #AIRegulation #ContentModeration npr.org/2026/08/30/nx-s1-594…

Nobody who ended up in an unauthorized deepfake thought it would happen to them either. That is the pattern almost every case follows. Someone with a public presence assumes the risk is far off. Then one day it is not. It is on their feed, tied to a product they never endorsed. By the time that happens, the tools that help most are the ones already set up. Takedowns still work. Detection still runs. But the people who saw the fake in the meantime have already formed impressions. Setting up your protection before anything happens is what keeps you from ending up in that pattern. Learn more at PersonaShield.com. #CreatorEconomy #AIProtection #DigitalRights #DeepfakeProtection

You do not own your face the way you own a photograph. That is where most of the confusion about likeness rights lives. A photograph has an author, a copyright, and a takedown path. A face has none of those things. What exists instead is a set of separate legal rights. Copyright protects the work you make. Trademark protects a brand if you have built one. The right of publicity protects your identity itself. Most people conflate the three, which starts to matter when AI enters the picture. A generated image of you is not a copy of anything, but something new that happens to look like you. That is what the right of publicity was built to handle, and also what it strains hardest against. 🔗 Our latest article walks through what the right of publicity actually covers. #AIProtection #CreatorEconomy #CreatorRights #IPLaw personashield.com/blog/right…

Michael Caine licensed his voice to a research project at the University of York studying how well people can tell an AI-generated voice from a real one. The AI version was built by ElevenLabs. Caine has said publicly that he actively polices unauthorized uses of his name, image, and likeness. This project shows the other side of that same posture. He is choosing which uses of his voice are welcome, and for what purpose. What makes the story land differently is the shape of his participation. He picked the use case, the collaborator, and the boundary. That is what proactive consent looks like when the individual is the one making the call. #CreatorEconomy #GenerativeAI #AILicensing york.ac.uk/news-and-events/n…

PersonaShield is now on mobile. Setting your safeguards, checking flagged items, and generating official images no longer requires a desktop. You can do all of it from your phone. Most of what happens with your likeness happens when you are away from a desk. Mobile access closes that gap so you can respond in the moment. Get started at my.personashield.com.

Every creator's line sits somewhere different. Some are happy with fan art and want nothing to do with brand work. Others open it up further. Some decide not to enable creation at all, and that is a complete answer too. What matters is that the line is yours, and you draw it once instead of deciding case by case. Above it, fans create officially and you earn from every image. Below it, nothing gets made through PersonaShield, and if someone makes it elsewhere, that is what the Protection Hub is watching for. personashield.com #CreatorEconomy #LikenessProtection #ResponsibleAI #FanArt

Researchers have tracked at least 304 AI-generated accounts pushing political content on TikTok alone since January. Most were built around images of real public figures. That is only what has been caught. The AI deepfake problem is not new. Election season just makes it harder to miss. Political ads move faster than almost anything else on a feed. They are paid to reach specific people, timed to a specific date, and blend into content that already looks the same. And the laws that exist mostly protect candidates and officials, not the person who ends up in the ad. 🔗 Our latest article walks through where AI deepfakes are showing up in this year's campaigns and where the real gaps sit. personashield.com/blog/polit…

"When I post something on the internet, I'm giving consent to post that. I didn't consent for them to remove my bathing suit on a picture." That is Jessieanna Campbell, one of several women recently interviewed by USA Today about waking up to find AI-altered versions of themselves running online without their approval. Somewhere in the last three years, public stopped meaning visible and started meaning available. AI tools treat every photo you have posted as material they can use, whether you agreed to that use or not. Consent has a scope everywhere else. You agree to a thing, for a use, for a period. The scope is the whole point, and it is the one thing none of these tools were built to see. Source: usatoday.com/story/life/heal…

Our CEO Phillip Shoemaker wrote a piece for Cybersecurity Insiders this week making the case that generative AI has turned a person's face and voice into an attack surface. Identity security has spent twenty years answering one question. Is this really you? AI has introduced a second. Is this really authorized? Answering that second question means building consent and authorization into how a face and voice get used, the same way credentials and access already work. That is what PersonaShield does. 🔗 Full piece linked below. cybersecurity-insiders.com/y…

The current model for handling AI likeness misuse is reactive. Something gets posted, you find it, you file a report, and you hope the platform acts on it. Meanwhile the content is already out there. PersonaShield flips the order. Detection runs continuously against safeguards you define. Reports move automatically. Official fan creation happens inside the boundaries you approved. Your likeness is not something anyone else gets to build on without asking. It is yours. PersonaShield.com

YouTube, TikTok, Meta, and X have all been asked what they are doing about AI-generated likeness of the people on their platforms. Three of the four have built something, which is a step forward compared to two years ago. A step is not a solution. The deepfake problem has grown faster than any of these tools were built to handle. What they offer starts with finding what has already been made, which is a piece of the puzzle, but only a piece. Giving people a real say in how their likeness gets used, opt-in by design and not opt-out, is the bigger question. Our latest article walks through where the tools are today and what could come next. personashield.com/blog/what-…

Starting August 2, AI-generated content in the EU has to be marked and labeled as AI-generated. Chatbots and AI agents have to disclose that they are not human. Both requirements are now enforceable, with a 38-person team inside the EU AI Office in Brussels authorized to investigate compliance. If someone generates a fake of you targeting EU users, the content is supposed to carry a marker that identifies it as synthetic. That gives a viewer something they never had before, which is a clear signal that what they are looking at was not made by the person in it. It does not stop the content from being made. It changes what the person seeing it is supposed to know when they see it. The biggest cost of a deepfake has always been in the moment someone believes it is real. A marked fake is still a fake, but a viewer who knows it is fake makes different decisions with it than one who does not. globalgovernmentforum.com/eu…

Karen Flowers is a licensed cosmetologist in Virginia with about 90,000 YouTube followers who watch her for hair care tutorials. Earlier this year, those followers started seeing an ad in which she appeared to sell life insurance. She had never heard of the product. That is where the fake endorsement space has ended up. Not the flashy scams and celebrity investment schemes people picture, but everyday commercial content built around real creators, running on the credibility they spent years earning. Our latest article breaks down what fake AI endorsements are and how they are affecting the creator space. personashield.com/blog/ai-ge…

The Protection Hub shows you every AI image flagged for crossing the safeguards you set. Where it was posted, how closely it matches your likeness, and a direct link back to the original so you are looking at the actual post. From there, the call is yours. Approve it and PersonaShield files the DMCA request on your behalf and tracks it through to removal. Or leave it up. Nothing moves without your final say, because you should be the one deciding what counts as acceptable use of your own likeness. Learn more: personashield.com/protection

Seven of the nine top image-editing tools on Hugging Face can be used to generate non-consensual nude images of real people, according to a new investigation by nonprofit AI Forensics. Hugging Face's own policies prohibit this kind of content. In practice, according to the researchers, safeguards were "virtually" absent. The finding matters less for what it says about one platform than for what it says about the AI industry. Hugging Face is one example. Meta rolled back a feature this month for the same reason. Others are being investigated. Every one of these cases has the same shape. A platform writes a rule, then leaves the enforcement to reports filed after the harm has already spread. Policies matter, but policies are only as strong as the systems built to enforce them. wired.com/story/hugging-face…

A deepfake of yourself online is not something you have to just accept. Finding these images and videos is more doable than most people realize, and so is getting them taken down. There are free tools that show where a photo has traveled, reporting paths on every major platform, and legal options when a platform will not act. The catch is that manual detection cannot keep up with the volume anymore. AI content spreads faster than any one person can search for it. That is where platforms built for this kind of work start to matter. Tools like PersonaShield watch continuously and surface content as it appears, so the response is already moving before you would have found it yourself. If you want a full walkthrough on how detection and removal works, or where to start, check out our latest article. personashield.com/blog/how-t…

TikTok is testing an opt-in deepfake detection tool. Creators verify their identity through Jumio by uploading a government ID and a live selfie, then TikTok scans the platform for AI-generated content matching their face. Opt-in is a good start, and built-in detection helps the creators using it, especially after Meta's Muse Image showed opt-out sets the wrong default. The bigger need is cross-platform protection. A creator protected on TikTok or YouTube can still have their likeness appear elsewhere, and no single platform can watch the wider internet on its own. Verification is another area still evolving. Uploading an ID to a third party adds friction and a new custodian to trust. Tools like PersonaShield use face-based verification instead, where the likeness itself becomes the credential, keeping the trust closer to the person being protected. TikTok's tool is a meaningful step, but the next layer is protection that runs across every platform, not just where someone posts. gagadget.com/en/718984-tikto…

Every official AI creation on PersonaShield starts with the person at the center choosing to be part of it. Bob Burnquist is one of the first creators building an official AI experience through us, letting fans generate AI images of him inside the safeguards he sets. For anyone who opts in, it opens the door to something new. A way to extend a brand, connect with an audience, and stay involved in how their likeness gets used. Learn more: personashield.com/official-c…

A recent Surfshark study put the global cost of deepfake fraud at $3.7 billion, with nearly nine out of every ten of those dollars lost in the last 18 months. What started mostly on social media has spread into every channel people use to communicate. Phone calls with a familiar voice. Video calls with a fake executive. Messaging apps with a cloned family member. People rely on the faces and voices they recognize. When those can be cloned convincingly, that trust becomes the exact thing being turned against us. The strongest defense starts upstream. If a likeness cannot be used without the person's permission, the harm downstream never gets a chance to happen. surfshark.com/research/chart…