We publish the problem, the architecture, and the reasoning behind our work.
“Artificial intelligence is eroding one of the foundational assumptions of modern technological superiority: that knowledge can be reliably traced to its source.”
As generative models continuously blend vast datasets, proprietary algorithms, and machine-generated code, that opacity creates a critical provenance gap, obscuring how data is altered, ingested, and reproduced. Because the United States can no longer reliably verify where machine-generated intelligence originates or whether it incorporates compromised intellectual property, what began as a commercial copyright dispute is becoming a national security challenge.
The article argues that patent law, copyright, trade secrets, and export controls all assume information moves between entities in discrete, attributable transactions, an assumption machine learning has quietly dissolved. Its proposed response is architectural rather than regulatory: a Strategic Knowledge Infrastructure, a federated verification layer embedded across national digital infrastructure that treats data provenance as an essential component of defense.
Read the Complete ArticleAn ongoing series by Sue Ghosh Stricklett, progressively addressing five major conceptual domains, from the intellectual foundations of the discipline through its engineering, governance, national-security, and societal dimensions.
Every item below is published as a downloadable PDF for archival use. Categories are populated on an ongoing basis.
Long-form analysis developed alongside our research program.
Why the governing question for AI-era information systems is whether knowledge can be traced, not whether it can be restricted.
Read more →What it takes to move lineage from a documentation practice to an enforced property of the systems themselves.
Read more →Current patent filings, the technology families, and licensing opportunities.
Read more →The research areas defining the problems our engineering addresses.
Read more →The platform mechanisms, each with an explicit implementation status rather than a capability claim.
Read more →Contact us to be notified as new volumes and papers are released.