Structural invisibility
How systems produce, retain, or erase forms of information that do not fit their established categories and timings.
MAMI Theory is the developing research perspective created and published by Mami, a Tokyo-based independent researcher. It examines how institutions and technologies shape what becomes visible, what can be evidenced, and who is able to act.
Research-led.
Public-facing.
Applied carefully.
Internationally connected.
WHAT MAMI IS
MAMI is an independent research and advisory practice led by Mami. It is not a university, public authority, certification body, or staffed institute. The name “MAMI Theory” identifies the research perspective and the public work built around it.
Understand the structure of MAMI →ONE INDEPENDENT PRACTICE · THREE CONNECTED DIVISIONS
Conceptual work on structural invisibility, institutional recognition, AI-assisted decisions, and responsibility.
Open division →Public writing, accessible explanations, and a durable record of what MAMI is investigating and what remains open.
Open division →Bounded, written reviews of evidence, responsibility, and escalation in a defined institutional or AI-assisted workflow.
Open division →BUSINESS / ADVISORY · CORE CAPABILITY
MAMI provides bounded, written reviews of how evidence, human review, escalation, and responsibility are organized in a defined institutional or AI-assisted workflow.
Current market test: AI-Assisted Claims Decision Readiness Review · USD 3,750 fixed fee · 7 business days.
No production-system access · No claimant data · No confidential documents at the initial inquiry stage.
Request a Readiness Review →INPUT → REVIEW → ESCALATION → RESPONSIBILITY
PLAINLY STATED
The current offer is a market test, not the whole identity of MAMI. It is not a formal audit, legal opinion, model validation, compliance certification, or implementation project. It is a bounded external starting point for understanding one workflow before a review, governance discussion, or operational decision.
See the deliverables and illustrative specimens →LATEST PUBLICATIONS
View all publications →Toward a Meta-Architecture of Mind and Invisibility in Disability, Education, and AI Ethics
M01PROPOSED HEURISTICA Proposed Analytic Heuristic for Examining Structural Invisibility
M02CANDIDATE FORMALIZATIONAn Unvalidated Candidate Formalization for Examining Information Retention
M03NORMATIVE PROPOSALResponsibility at the Level of Epistemic and Institutional Design
M04How systems produce, retain, or erase forms of information that do not fit their established categories and timings.
How evidence, professional judgment, exceptions, and final responsibility are organized around AI-supported decisions.
How ethical questions can be examined at the level of institutional and epistemic design—not only individual intention.