About

I spent my career at Kansai Electric Power Company (KEPCO, Japan's regional power utility), then in my own independent business, then at Samsung SDS in Korea, and then retired. After that I taught electrical theory at a vocational training school for three years, and observed a vocational rehabilitation center from the inside. Now I run Mindseed Research as a sole-proprietorship initiative, pursuing a 20-year research program centered on Bonnō × AI.

Career

• KEPCO — design and implementation of instrumentation panels and relay control panels. The discipline of power-system engineering is something I learned with my bones
• Independent business — design and manufacture of sequence control and quality-inspection equipment for automotive production lines and the like. The rigor of quality on the manufacturing floor is something I learned here
• Samsung SDS — infrastructure engineering. Two philosophies have stayed with me since: "mature technology is safety first" and "mismatch is the best fit"
• Vocational training school, 3 years — after retirement, teaching electrical theory. My top priority was "no one falls behind"
• Vocational rehabilitation center — I observed disability employment-support services from the inside, and came to understand that people with disabilities have a real inner strength of their own
• Now — independent as Mindseed Research, beginning a 20-year research program

Within the bounds of professional confidentiality, I do not disclose the specifics of my technical work at KEPCO or Samsung SDS; I record only my role and the influence it had on how my thinking took shape.

Technical background

For more than 40 years I have done infrastructure engineering centered on C/C++ and assembly. I have watched the technological transition from 8-bit CPUs to today's generative AI from the field.

The integrated stack I am using now is PHP/MySQL (existing assets), Flutter/Firebase, Next.js, React Native, and Claude / ChatGPT / Claude Code. I hold to my infrastructure-engineer philosophy of "use mature technology at the necessary and sufficient scale," while absorbing the new development practices of the AI era.

Important life events (self-disclosure)

To keep transparency as a researcher's narrative, I want to record here three events that shaped the direction of my life.

The BBS case of the 1990s

Around age 30, I ran a grassroots Bulletin Board System on SunOS — 9 modem lines, about 300 members. Members uploaded obscene images and other members downloaded them, and as the operator I was held responsible for the conduct of the service. I was arrested by 5 detectives, defended pro bono by 6 lawyers, and lost the case. The verdict became the first Japanese precedent in which "a hard disk was deemed to constitute the public display of obscene material." The sentence was 3 years suspended.

That experience was how I came to know the boundary of technology, law, and society from the inside. It is also the starting point of how I came to define myself as "the opposition party of the IT industry."

The failure of CineBASIC

I spent half a year developing "CineBASIC," a new language for video-generation AI, and gave up. I was defeated by the difficulty of controlling the inherent fluctuation of generative output. But through that experience I came to understand, bodily, that the "fluctuation of generative AI" is not a bug or a tuning issue but a fundamental problem of training data and evaluation design. One of the reasons my research program stands on Economies of Precision is the inside knowledge I gained from this defeat.

A close relation's victimization by fraud (May 2026)

Someone close to me was defrauded of approximately 7 million yen (about $47,000) in an FX-investment scam. That event was what made me take the long-incubated anti-scam vision into full-scale work, formally make Mindseed Research independent, and start the research program. The starting point of my philosophy of "Protect Your Only Life" lies in that lived experience.

Core philosophical commitments

• Protect Your Only Life — I frame scam victimization not as financial loss but as a violation of the singular life itself
• Mismatch thinking — unexpected combinations across disciplines yield the strongest market differentiation. A sales philosophy I picked up during my Samsung SDS years
• The opposition party of the IT industry — where the giants go wide and shallow, I go narrow and deep. The strategic positioning of Economies of Precision
• 20-year span and successors — not short-term outcomes, but a research program left in a form that someone can inherit

Replies are welcomed from

• AI Safety / Alignment researchers — at international AI labs, in the LessWrong community
• Cognitive science / neuroscience researchers — at universities and research institutes in Japan or abroad
• Buddhist scholars — at institutions such as Bukkyo University, Komazawa University, Ryukoku University, and others
• Researchers in the victimology of fraud — domestic and international
• Successor candidates: early-career researchers and graduate students — those interested in a 20-year research program

Channels: matsuura@pyol.net, or via the contact page. I cannot promise to reply to everything, but I read everything.