KINETK is a tech company that tracks content across the internet using invisible watermarking, AI fingerprinting, and blockchain registration. Born from a July acquisition by Trips, an Intellectual Property (IP) tokenization company, of Mentaport, an IP protection and content tracking platform, this new entity enables creators and IP holders to see exactly where their content appears online, even when it has been modified or embedded within other works.
The story of KINETK begins with a serendipitous meeting between Trips founder Michael Finkelstein and Mentaport co-founder Mariale Montenegro, who had been working on different aspects of the same problem.
“We got introduced through a mutual venture capitalist association,” Michael recalls. “We sat down for coffee in late October and within 45 minutes agreed to work together.” Their skills complemented each other. “Michael is the business person I need, and I’m the technical person he needs,” Mariale says.
Michael brought experience from building companies in music and fintech, including The Credit Junction, a fintech lending and data analytics company he founded and led from 2013 to 2019. Mariale contributed her expertise in computer vision and AI from her time at Magic Leap and Apple, where she focused on developing tools for creators.
Based in New York and San Francisco, KINETK serves both individual creators on platforms such as YouTube and TikTok, as well as large enterprises with major IP portfolios.
“We want to flip the script on the model where [today] people are focused on where IP starts, where it gets posted, which platform it goes on,” explains Michael, now serving as CEO of KINETK. “We want to focus people on where it gets consumed and where it goes afterward, because that is the unlock. When you think about intellectual property, it’s far bigger than the creator economy. It powers everything.”
Three-Layer Tech Stack Powering Attribution
KINETK’s system begins with two parallel processes applied to each piece of content: embedding an invisible watermark directly into the content and creating an AI fingerprint based on the content’s characteristics.
“When we get a piece of content, we do two things,” explains Mariale, who serves as the company’s CTO. “We will embed an invisible watermark in the piece of content or an AI fingerprint inside the piece of content.” The watermark functions as a discreet marker embedded within the content itself, while the AI fingerprint serves as a unique identifier generated from analyzing the content.
Once the content has been processed, KINETK registers its metadata on the blockchain, creating an immutable record of ownership that serves as the foundation for all subsequent tracking.
“We are using blockchain because it’s the best technology to do this,” Michael emphasizes. “The reality is blockchain is meant for, we believe, the immutable ability to record facts on a ledger.” These facts include not just ownership, but other critical metadata about the content that establishes its origin.
The third component of the system consists of what the company calls “agentic AI scanners” – autonomous agents that continuously scan the web looking for the invisible watermarks and AI fingerprints KINETK has embedded. Mariale explains their unique approach: “The scanner never knows what it’s seeing. It’s just looking for that invisible code for us to be able to detect it.” When a scanner identifies a registered piece of content, it reports back to the blockchain, which confirms ownership and notifies the creator through KINETK’s dashboard.
Beyond Copyright Protection
KINETK differs from traditional content protection systems in its ability to track content that has been modified, embedded within other works, or fragmented. “Content is going to move,” Michael says. “Pieces of it are going to move across the web in ways we haven’t seen before.”
For creators, this visibility can potentially lead to financial benefits. “If you think about an influencer on TikTok, maybe they’re charging $1,000 a post for the brand deals,” Michael explains. “They’re doing that based on where the post is on TikTok. They’re not actually understanding that their post may circulate elsewhere.”
With data about their content’s true reach, creators can leverage this additional exposure in their brand deals or identify new revenue opportunities.
Making Complex Technology Accessible
Despite the technology working behind the scenes, KINETK has prioritized creating a streamlined user experience that requires minimal technical knowledge. “Within three clicks, the magic happens,” Michael says of their platform.
The user journey begins when creators either connect their existing channels (currently, YouTube or TikTok) to KINETK or upload content directly to receive protection. The system automatically processes this content, embedding the necessary markers and registering ownership.
“You don’t even know it’s running on the blockchain. You don’t even need to care,” Michael notes, emphasizing how the technical complexity is hidden from users.
Once their content is protected, creators gain access to a customizable dashboard that displays instances where their content has been detected elsewhere on the web. Mariale explains how users maintain control over the information they receive: “You can have filters. For example, you can see if your content appeared on TikTok, or if it appeared with more than 10 likes.” This prevents notification fatigue while ensuring creators see what matters most to them.
When KINETK detects a creator’s content, it presents multiple response options. “A creator’s first point of order is the information we want to give them,” Michael says. “For instance, ‘Did you know these ten links contain your content?’”
Creators can then choose to do nothing, issue an automated takedown request, or potentially capture revenues if their content appears on platforms that monetize it, though the specific mechanisms for revenue capture are still being developed.
The Process of Content Tracking
Before KINETK, Mariale was an AI lead at Magic Leap, developing key tools and algorithms for spatial computing. At Apple, she focused on creating tools and applications to empower all Apple users as creators. This experience provided her with insights into how to approach the challenge of content tracking at scale.
“I got obsessed with the idea that we all have a brain and we’re all the same biological creatures, but so different at the same time,” Mariale explains, describing her early fascination with computer science. “Can we make the computer see the world, react to the world differently depending on who’s using it?” This question led her to work with computer vision, AI, and deep learning – technologies that now form the foundation of KINETK’s tracking capabilities.
Building a system capable of scanning the entire internet for specific content signatures presented major technical challenges. Rather than attempting to understand what content means, KINETK’s system focuses solely on identifying unique signatures. “The Agentic scanners look for that watermark and AI fingerprint,” Mariale explains. “When we find something, we don’t say, ‘Oh, we found content that’s a video of a cat.’ We say we found ID 1, 2, 3, etc.”
As the founders point out, this signature-based approach enables KINETK to operate at scale while maintaining privacy and minimizing computational overhead. It also means their system can work with any type of digital content – images, videos, audio – using the same core technology.
Why Content Attribution Matters
The launch of KINETK comes at a time Michael and Mariale identify as a pivotal moment in how content is created, distributed, and monetized. The growth of AI tools has changed the creative field, making both content creation and tracking more complex than before.
“IP used to reflect mostly physical assets,” Mariale says. “When the internet started, we began to notice a shift in IP. Streaming took off, and platforms like MySpace and YouTube started allowing people to express themselves uniquely. This started the content creator revolution. An internet of read and write.”
AI has changed this dynamic. “With AI, that just takes all those roadblockers away,” Mariale continues. “You don’t know how to edit. Don’t worry. You don’t have a studio to record a song. Don’t worry. You don’t want to show your face. Don’t worry.” The result has been an increase in content creation that makes attribution both more challenging and more essential.
Michael notes that this shift has raised the importance of intellectual property in public consciousness. “Fifteen years ago, no one cared about IP. No one knew what it was,” he says. “Over the last year, it’s been on the front page of every mainstream journal article, and it’s on the lips of everyone talking about intellectual property.” This increased awareness reflects a growing recognition that digital creations hold real value that needs protection.
A Vision of an Ecosystem
In the near future, KINETK plans to expand beyond its initial tracking capabilities to foster a broader ecosystem for content attribution and protection. One key initiative currently in development is the Sentinel program, which Michael describes as “an advanced ambassador program that will allow anyone to help engage in this ecosystem.”
According to him, this program will enable KINETK users to participate in identifying protected content across the web, extending the company’s reach beyond what its own scanning technology could achieve alone.
Mariale elaborates on how this community approach will work: “If you’re able to find the content and flag it with the help of the tools we’ll be providing, then you can also be recognized.” This creates a cycle where creators help protect each other’s work while receiving acknowledgment for their contributions.
The Sentinel program shows KINETK’s broader vision of creating not just a tool, but an entire ecosystem for content attribution. “We are excited about allowing anyone who’s a part of the KINETK Network not only to track their own content, but [also] participate in ways that help other content creators,” Michael explains.
The company plans to launch its platform to early access users in September, marking the beginning of what its founders hope will be a new phase in content attribution.
“We want to make sure that whatever you see, no matter where and what platform it is, you know where it came from, who made it, what it is from,” Mariale summarizes. “When it gets consumed, we’ll let you know and we’ll make sure the revenue goes back to you, wherever it is, even if it’s been stripped, manipulated, or whatnot.”