Information Sovereignty

My foundational work focused on the intersection of biological anthropology and legal property rights. In my Columbia University thesis, "Genetics and the Consumer," I argued that DNA is a unique category of property that cannot be truly "abandoned." I identified critical regulatory gaps where "biological exhaust"—the digital data extracted from physical samples—was treated as public property rather than a sovereign asset. 

As autonomous systems began to scale, I identified a parallel between biological genomes and artificial neural networks. I pivoted my research to treat AI models not as static code, but as dynamic organisms that evolve through training. By applying principles of computational neuroscience, I began engineering the mathematical frameworks necessary to treat model weights as "Neural DNA." 

 My current research provides the technical and mathematical safeguards for institutional model sovereignty. This era focuses on Forensic Auditing—mapping complex neural audit trails to emerging global AI governance legislation, such as the EU AI Act and NIST frameworks.

In my framework, AI weights are not just parameters; they are Neural DNA. Just as genetic markers define biological property, model weights and state transitions define institutional property. My research ensures that as models evolve through training, their intellectual lineage and proprietary sovereignty remain legally and forensically intact.

• Information Integrity: Using Shannon Entropy to quantify the informational "purity" of training data and model outputs.
• Neural Homeostasis: Implementing "Homeostatic Governors"—autonomous circuit breakers based on inhibitory veto logic to prevent model drift and catastrophic over-excitation.
• Data Provenance: Maintaining a cryptographic chain of custody for foundational data, ensuring the purity and sovereignty of the model’s "genome".