How it works
Separate technology, business, and governance issues
Mixed AI, security, governance, and risk topics are separated into decision points, review points, operating roles, and next actions.
Issue structureDecision pointsReview points
About
An independent practice for decision-ready material across AI, security, and technology risk.
Fragment Practice turns mixed issues around AI adoption, security governance, internal regulations, and technology risk into material organizations can use for decisions, explanation, review, and next-step work.
The practice connects experience across system development, cybersecurity, IT risk, governance, enterprise advisory, AI governance, and management-facing explanation material.
Practice overview
Clarifying decisions, responsibilities, and operating models even when the work is already under way.
In AI, security, and governance work, technical, operational, and management concerns arrive mixed together. Fragment Practice treats them as one connected question — what to promote, where control is needed, who decides, who reviews, and what to record — rather than as separate tool, compliance, or implementation tasks.
That question becomes usable material: issue memos, decision material, role and operating design, review points, and handoff-ready next actions that a team keeps using after the conversation ends, and can carry into decisions, reviews, and operation on its own.
Founder profile
Yasuhiro Shinsho
Founder, Fragment Practice LLC
Yasuhiro Shinsho founded Fragment Practice after building and assessing systems at BIPROGY, consulting on IT risk and security at KPMG Consulting, and advising on security and cloud / AI risk at Nomura Research Institute / NRI Secure Technologies. The work moved from development and assessment, to structuring risk and control, to preparing what management and stakeholders need in order to decide.
Today he works on the questions organizations stop at: whether customer information may be entered into a generative AI service, how far to go in meeting a security requirement a customer has asked for, who reviews a new external service and who is answerable once it is in use. Across AI governance, security governance, internal regulations, and technology risk, those questions are turned into decision material, review points, responsibility boundaries, and role / operating design.
Since becoming independent, he has supported AI governance policy structuring, human/AI role and operating design, AI-enabled service concept review, security requirement structuring, third-party review material, and handoff material for enterprise contexts.
Takamatsu, JapanIndependent advisoryDecision / operating model
Topics for talks and interviews
From professional work
- AI governance and organizational decision-making
- Responsibility boundaries and review between people and AI
- Security governance and technology risk
- Turning AI and security questions into material management can decide on
From the founder's own experience
- How working with AI changes the way one person works
- Running an independent practice from a regional city
Talks, lectures, panels, and interviews can be requested through Contact.
Ask about a talkElsewhere
Official information
Legal contains official business information, service and product boundaries, licensing scope, payment and delivery basics, refund baseline, and general commercial terms.
Base and support style
Takamatsu-based, remote-first advisory support.
Fragment Practice is a Japan-based practice led by its founder, working remotely with organisations in Japan. English-language engagement is available for overseas parent companies and stakeholders who need to review the same material in English. In-person interviews, workshops, or important meetings can be discussed when they are useful for cross-functional alignment, management explanation, or major review points.
Takamatsu-basedRemote-firstIn-person when useful
Business information
Registered address, corporate number, qualified invoice issuer registration, and other business information are listed on Legal.
Fragment Practice LLC / Takamatsu, Kagawa, Japan
How the practice works
Focused on structuring judgment, responsibility, and usable next-step material.
Fragment Practice clarifies issues, roles, review points, and responsibility boundaries, without taking on implementation or operations itself. The aim is to leave internal teams with less ambiguity about what to decide, how to explain it, where to review, and what to hand off.
How it works
Keep promotion and control in the same discussion
AI and security topics are organized so that adoption, control, responsibility, and practical operation can be discussed together.
AdoptionControlPractical operation
How it works
Clarify roles across business, IT, security, risk, and vendors
The work clarifies what each stakeholder should decide, review, explain, operate, or hand off.
Role boundariesStakeholdersHandoff
How it works
Set how far work is delegated and where people review it
For people, AI, automation, and vendors alike, the work defines how far each is trusted to act, where a person reviews, and on what condition to stop and return the decision.
Delegation limitsReview and stop pointsExceptions
Who does what
Clarifying the roles of deciding, operating, and building
Management decides acceptable risk, investment, and key policies; business and control functions design the operating conditions and reviews; internal teams or vendors design and build to those conditions. Fragment Practice prepares the decision material, review points, responsibility boundaries, and handoff conditions needed in between. Each holds a different responsibility; this is not a hierarchy.
ManagementBusiness and controlImplementation teams
Next step
Start from the current situation.
To take a current situation further, start from Contact. Services covers the engagement formats, and Cases shows similar patterns of work.