Alexander N. Cutler

A little about me.

Never really thought of myself as a consultant or a “founder”: just a guy who would rather be called wrong and be wrong than called right and be wrong.

I started my first company at 26, fresh out of school, with no partners, no inheritance, and a questionable degree of enthusiasm about Truth with a capital T. I never mistook “this is how I've seen it done” for “this is how it should be done,” so I set out to solve problems others were merely managing.

But those problems were never just puzzles to solve. From the start, it has always been the people behind the invoice—and what's at stake for them—that I ultimately serve. I believe it’s that sincerity and assiduousness which have differentiated my firm and grown it into a globally active consultancy spanning several domains of regulated industry.

At some point, the University of Washington caught wind of my antics; they appointed me as an affiliate faculty professor in the Biomedical Regulatory Affairs M.S. program and invited me to advise the Vice Provost for AI via the AI@UW Advisory Council. I hadn’t applied for either. Increasingly, to my genuine surprise, I'm also being invited to speak at world-class institutions such as Fred Hutchinson Cancer Center and to be interviewed on podcasts.

I've seen some of the worst damage in regulated industry come not merely from Scrooge McDuck executives disregarding compliance—as the public often assumes—but actually from “credentialed” professionals assuring those same executives that broken controls work. So I invented the patent-pending Governed Regulated-Enterprise Simulation System (GRESS) to give business leaders auditable evidence of their professionals' judgement before costly failures expose patients and capital to risk. Yet I soon discovered my invention might have implications far beyond human performance assessment: the same system could systematically test the judgement of machines themselves, opening a new empirical approach to one of the defining challenges of our time—the alignment problem.

My most consequential work is my forthcoming book, Glyphic Engines: The Machine Class for Coherence. It shows how the unstable category of “AI/SI” has exposed a hitherto uncharacterized machine class capable of sustaining meaningful coordination without anthropomorphic notions of consciousness or agency. Glyphic Engines presents far-reaching implications for cybersecurity, military coordination, and beyond—even challenging the assumed divide between what is “synthetic” and what is “natural.” In fact, an early preprint of my work, released under its original title, independently and precisely anticipated the kind of cyberattack enabled by cross-machine coordination mechanisms later observed in the OpenAI–Hugging Face incident nearly two months before OpenAI publicly disclosed its technical findings.

If you read one thing I’ve written, make it Glyphic Engines—coming in 2027.