--- title: "About XI Objects" description: "About XI Objects Inc." url: https://xiobjects.com/about source: XI Objects --- # About XI Objects AI-generated content is everywhere. Knowing what's real, where it came from, and whether you can trust it is a different problem entirely. XI Objects exists to solve that problem. ## Where It Started XI Objects grew out of a practical need. Isaac Livingston, Principal Technologist and co-founder, was building AI systems for real business use and kept running into the same wall: there was no good way to feed structured, trustworthy data into RAG pipelines and have that trust carry forward into the model's response. So he built one. **XION** (XI Object Notation) is a format Isaac invented that works as a Markdown Object Notation language. XION artifacts are human-readable and machine-parsable by design. Plaintext fits naturally into how LLMs consume information, and the structure means no transformation step is needed for parsing. When Isaac started asking how to protect the integrity of that plaintext, a PKI architecture followed directly from the requirements. **Kenneth Cavner**, Director of AI Practice and co-founder, holds a Master's in Mathematics and has deep experience in AI systems. When Ken saw what Isaac was building, he recognized that XION wasn't just a transport format. It was an attribution substrate. Used correctly, it could reduce hallucinations, improve data freshness, and create a verifiable chain of custody for information flowing into AI systems. That's where XI Objects as a company started. ## The Analog Hole Expanding XION's scope from data to media meant solving a hard problem: how do you identify and attribute an image or video that's been photographed off a screen, re-encoded, filtered, or otherwise transformed? No embedding survives that. Watermarks don't either. The industry calls this the **Analog Hole**, and nobody had a satisfying answer. Isaac's approach started from an unusual place. Rather than treating an image as bytes, he asked a simpler question: if a person can look at a picture and recognize what it is, that recognition is based on perception, and perception is based on light. Working from first principles in visual perception research, he built a system grounded in how recognition actually works. The result is a two-stage perceptual fingerprinting system. Stage one generates a spectral fingerprint used as a fast index for candidate lookup via vector distance search. Stage two applies heavier discriminatory processing to the candidate set for precise identification, including across platform transformations that destroy conventional signatures. Ken extended this to video through his foundational research in Luminance Waveform Analysis. When they compared notes, Isaac observed that light and sound are both wave phenomena, and that the same mathematical framework should apply to audio. Adding a third dimension to the image processing model through horizontal and vertical scanline analysis bridged the gap. Audio attribution followed from there. The system works. When they shared early results with a colleague familiar with C2PA and the Analog Hole problem, that colleague arranged a warm introduction to Microsoft. The conversation that followed validated what they'd built. ## The Institute of Provenance The commercial product is XI Objects. The standards and trust network live in a separate, independent entity: the **Institute of Provenance**. The Institute is an incorporated organization with formal bylaws and a non-profit attorney engaged for 501(c)(6) registration. Its governance model makes AI provenance infrastructure trustworthy at the institutional level, not just the technical one. A provisional patent covering the core technology includes over 40 claims. The trust network runs on hierarchical PKI with distributed Orbital nodes. It's designed to bring the XION standard and attribution infrastructure to regulated industries where AI credibility isn't optional: healthcare, finance, legal, government. Sectors where "it usually works" doesn't cut it. ## The Founders ![Kenneth Cavner](/assets/founders/kenneth-cavner.jpg) ### Kenneth Cavner Co-founder & Director of AI Practice 30+ years in AI systems, statistical modeling, and applied mathematics. Master's in Mathematics. Inventor of Luminance Waveform Analysis (LWA). ![Isaac Livingston](/assets/founders/isaac-livingston.jpg) ### Isaac Livingston Co-founder & Principal Technologist 25+ years in technology spanning enterprise infrastructure, cloud architecture, AI systems, and full-stack engineering. Inventor of XION.