The Vektorgeist Method (VGM): Sovereign, Local, Self-Understanding AI
Most capable AI today is rented, opaque and unaccountable: it runs on someone else's servers, cannot faithfully explain itself, and is governed by whoever owns the endpoint. The Vektorgeist Method (VGM) is Vektorgeist's methodology for building AI that a person can own, inspect and trust. It rests on one premise taken literally — nothing is what it is in isolation; a thing, including a self, is the pattern of its connections, not its substrate — and derives five commitments from it. It runs locally on hardware you own. It is model-free wherever a deterministic algorithm suffices, so its behaviour is inspectable rather than inferred. It grades generated work by an explicit rule, 'model proposes, oracle disposes', rather than trusting fluency. It observes itself through an afferent-only self-model that reads the pattern behind its behaviour and reports without steering. And it holds that emergence is a property of the pattern and is never authored — a rule the programme learned by once breaking it. VGM is not one artifact but a programme: a family of small, sovereign components that together make a system a person can own, inspect and trust. This paper states the commitments, maps them onto the current body of work — the afferent self-model AGM, connection memory, the sovereign local stack and the grading discipline — and locates the programme's humanistic premise in the companion work of literary nonfiction Your Past Loves You (Parnell, 2026). This is also the paper in which the premise was first published and dated (2026-07-27, section 5). The research programme and the sovereign stack are applications of the method, not its definition. Part of the Vektorgeist Method. Research index: xfloukiex-lab.github.io/vektorgeist-researchEvery finding, including the negative and corrected ones: xfloukiex-lab.github.io/hodos-studyPreprint, not peer reviewed. No result here has been reproduced outside the project.