Below is our recent interview with Kosuke Shirako, the Founder of SHIRO & Co.:
Q: How has the Kosuke Protocol evolved through multiple public implementations?
A: The Kosuke Protocol originally began as a method for connecting fragments — signals, observations, changes in behavior, institutional shifts, and small anomalies that did not yet have a clear name.
Its basic sequence is:
Fragment → Observation → Connection → Meaning → Form
What changed through the public implementations is that “Form” became much broader than I initially expected.
At first, I thought the output would mainly be concepts, research notes, or business hypotheses. But once I started applying the same protocol repeatedly, the outputs began to diverge.
Climate Temporal Compression became a way of interpreting climate change through usable human time.
The Supervisory Layer became a governance and runtime-supervision concept.
Observation-to-Proposal became a working business system that connects structural observation with company-specific proposal development.
LIFE became fiction and film.
So the Kosuke Protocol is no longer only a method for interpreting fragments. It has become a reusable transformation structure that can move from observation into different kinds of form depending on what the observation requires.
The important point is that the output is not predetermined.
The protocol does not begin by deciding, “This should become software,” or “This should become a report.”
It begins with observation and lets the form emerge later.
Q: How can the same protocol produce different forms — research observations, supervisory concepts, business systems, and narrative work?
A: Because the protocol is concerned with the structure underneath the material, rather than with the final medium.
For example, a market signal, a customer conversation, a climate pattern, or a fictional scene may look completely different on the surface.
But the same questions can still be asked:
What changed?
What is the structure behind that change?
What connects this fragment to other observations?
What does the change mean?
What form should that meaning take?
That last question is where the outputs diverge.
If the observation reveals an operational problem, the form may become a business system.
If it reveals a missing institutional function, it may become a concept such as the Supervisory Layer or HOLD.
If it reveals something difficult to express analytically, fiction or film may be the better form.
LIFE is a good example.
The underlying observation is not simply “AI will change society.”
The question is: what remains human when AI works correctly?
A report could explain that analytically, but a short scene can make the same structural change felt in a different way.
So I see software, research, concepts, and narrative as different output layers of the same observational process.
Q: What did you learn from applying the protocol to Climate Temporal Compression, The Supervisory Layer, Observation-to-Proposal, and LIFE?
A: The biggest lesson was that the same protocol can expose very different kinds of missing structure.
Climate Temporal Compression showed me that climate change can be observed not only through temperature, but through the shrinking amount of time in which human activity remains practically usable.
That shifted the question from:
“How hot is it?”
to:
“How much usable time remains?”
The Supervisory Layer revealed a different missing structure.
As AI systems become more autonomous, monitoring alone is not enough. There also needs to be a formal ability to prevent execution from proceeding while important conditions remain unresolved.
That is where HOLD emerged.
Observation-to-Proposal taught me that observation should not automatically become action.
Once an observation enters a business workflow, provenance, permission, uncertainty, and human judgment still need to remain visible.
That led to the Handoff Boundary and Proposal Boundary.
LIFE taught me something else: some structural changes are better understood through narrative than explanation.
A fictional scene can hold ambiguity without forcing an immediate conclusion.
Across all four projects, I learned that the protocol is most useful when it does not rush toward an answer.
Its value often comes from identifying the missing layer between what already exists and what society, organizations, or people are beginning to need.
Q: What remains constant across these very different implementations?
A: Three things remain constant.
First, observation comes before solution.
I try not to begin with a product category, technology, or business model. I begin with something that appears to be changing.
Second, facts and interpretation must remain distinguishable.
This is especially important when working with AI.
A source fact, an observation, and an inference are not the same thing.
The recent Observation-to-Proposal implementation makes this explicit by preserving provenance and separating what was stated, observed, inferred, verified, or excluded.
Third, the form must remain open until the meaning becomes clearer.
That is probably the most important part of the Kosuke Protocol for me.
The same observation may eventually become a concept, interface, business proposal, film, educational experience, or something else entirely.
What remains constant is not the output.
What remains constant is the movement from fragment to observation, from observation to connection, from connection to meaning, and only then to form.
Q: What do you think the Kosuke Protocol is becoming now?
A: I think it is becoming less like a framework and more like an operating architecture for turning observation into different forms of action without collapsing the distinctions in between.
Earlier, I described it mainly as:
Fragment → Observation → Connection → Meaning → Form
I still think that sequence is correct.
But the public implementations have added another dimension.
There are now boundaries between those stages.
An observation may be valid but not yet usable.
A signal may be meaningful but not yet actionable.
A proposal may be plausible but still need to remain on HOLD.
A fictional observation may be important precisely because it should not be converted into a recommendation.
So the Kosuke Protocol is increasingly becoming a system for managing the transition between observation, interpretation, permission, and form.
I do not think its final identity should be fixed too quickly.
One of the reasons it has produced such different outputs is that I have tried not to define it too narrowly.
For now, I would describe it as a reusable structure for observing change, identifying what is missing, and deciding what kind of form — if any — should emerge from that observation.
That may be a business system.
It may be a supervisory concept.
It may be a film.
The protocol should not know the answer in advance.
