Open Business Protocol

An open protocol for describing, organizing, and evolving any business system.

For human operators and AI agents.

First Published: 2026-04-01

I.

Why does business appear complex?

The apparent complexity of modern business is not inherent to business itself. It is an artifact of fragmentation — of systems that cannot speak to one another, each defining "customer," "task," or "value" according to its own private vocabulary.

What is missing is not more software. It is not a better API. What is missing is a grammar — a shared structure prior to any implementation.

II.

Four bases. Everything else is a sequence.

being — A —

The Subject. Will.

resource — T —

The Object. Matter.

flow — G —

The Process. Time.

context — C —

The Frame. Why and where.

III.

A protocol, not a product.

ISO does not design products. It defines the conditions under which products can be manufactured and exchanged with confidence across institutional boundaries. OBP occupies an analogous position for business operations.

Universal

The grammar does not assume an industry, a technology, or an era. A trading house in 1850 and a network of autonomous agents in 2050 are both fully describable within the same four primitives — at different resolutions, but without loss of coherence.

Generative

The protocol is a grammar, not a list. Any business activity, in any industry, at any scale, can be derived from its primitives.

Implementation-agnostic

An OBP-conformant description does not encode the assumptions of any platform. A reference implementation exists as proof of feasibility, not as a constraint on conformance.

IV.  Whitepaper

OBP v1.0: The Grammar of Business as a Living System

Open Business Protocol — 2026-04-01

Abstract

Open Business Protocol (OBP) is a universal protocol for describing business domains through four irreducible primitives: being, resource, flow, and context. Implementation-agnostic by design, OBP establishes a shared grammar through which any business system — human or machine, small enterprise or large organization — can be coherently described, organized, and evolved. This document presents OBP v1.0: its philosophical foundations, structural principles, and implications for the age of intelligent agents.

Introduction: The Problem of Complexity

Why does business appear complex? When we examine the operational architecture of any organization, we encounter layers of specialized software, proprietary data structures, and siloed workflows. Each tool speaks its own dialect. Each platform defines "customer," "invoice," or "task" according to its own schema. The apparent complexity of modern business is not inherent to business itself — it is an artifact of fragmentation.[1]

But the problem is not new. Organized human activity has always generated coordination challenges that exceed the capacity of any individual to manage alone. Every era produced ambitious responses: Pacioli's formalization of double-entry bookkeeping in 1494 gave merchants a shared grammar for recording value across time and counterparty;[6] scientific management at the turn of the twentieth century attempted to reduce human labor to transferable flows;[7] enterprise resource planning systems encoded entire organizational logics into software that could be administered at scale. Each was a genuine achievement. Each resolved a portion of the problem. And each was overtaken, in time, by the complexity it could not anticipate — as the confident enterprises of 1928 discovered when the foundations gave way beneath them.

AI agents are not the origin of the current crisis. They are its latest accelerant.[3] The proliferation of autonomous systems — each capable of acting, deciding, and transacting on behalf of the organizations that deploy them — has compressed the timescale in which fragmentation becomes critical. What once took decades to calcify now hardens in months. The same incoherence that once frustrated human operators now misdirects autonomous processes at machine speed.

Existing frameworks have addressed symptoms rather than roots. Low-code platforms offered visual abstractions that masked underlying incoherence. AI agents have introduced new capabilities without resolving the fundamental problem: there is no shared language for describing what a business is or does.

This is not merely a technical problem. It is a linguistic one. OBP proposes a solution at the level of grammar itself.

The Four Bases

Every grammar rests on a finite set of irreducible elements. OBP proposes four.

being — The Subject (Will).  Being denotes any entity capable of intention — human, artificial, or organizational. A being is not defined by its substrate but by its capacity for directed action. It can configure its environment, coordinate with other beings, and reflect on its own behavior. Being is the locus of sovereignty: the origin point from which all meaningful action flows.

resource — The Object (Matter).  Resource denotes anything that a being can own, transform, consume, or create — money, data, intellectual property, physical goods, relational capital. A resource without a being to ascribe meaning to it is merely raw matter. The relationship between being and resource constitutes the primary axis of ownership and value creation.

flow — The Process (Time).  Flow denotes the structured description of change over time. A flow is not merely a sequence of actions; it is the encoding of how a being operates on resources within a given context. Flow is always contextually bound: the same sequence of operations produces entirely different outcomes depending on the frame within which it is executed.

context — The Frame (Why/Where).  Context denotes the conditions of meaning — the market, the regulatory environment, the organizational purpose, the emotional state of participants. Context is not background noise; it is the interpretive frame that gives flow its direction and resource its value. Without context, flow becomes purposeless motion.

These four are not arbitrary. They emerge from a structural analysis of what any intelligible account of business activity must include. Remove any one and the description collapses: without being, there is no agent; without resource, no subject matter; without flow, no change; without context, no meaning.

The relationship between these bases is not merely additive. They form two complementary pairs, each exhibiting a distinct coupling character.

Being and resource constitute a fluid pair (analogous to the adenine-thymine base pair in DNA). A being freely acquires, transforms, and releases resources. The coupling is dynamic: ownership changes, value transforms, relationships evolve.

Flow and context constitute a structural pair (analogous to guanine-cytosine). A flow cannot be understood independently of its context. The same process executed in different markets, under different regulations, or within different organizational cultures produces fundamentally different outcomes. This coupling is strong: it is not possible to fully specify a flow without specifying the context in which it operates.

The structural analogy to DNA is intentional but bounded. We invoke it not to reduce business to biology, but to suggest that just as biological life is encoded through a small number of base pairs capable of infinite combination, business activity may be fully described through a small number of conceptual primitives capable of generating the entire observable complexity of organizational life.[4][5]

The Protocol as a Universal Grammar

The International Organization for Standardization does not design products. It defines the conditions under which products can be designed, manufactured, and exchanged with confidence across institutional and cultural boundaries. ISO 9001 does not specify what a company should make; it specifies how quality management processes should be structured so that the output is consistently reliable.

OBP occupies an analogous position in the domain of business operations. It does not specify what a business should do, what tools it should use, or what markets it should enter. It specifies the grammatical structure through which any business activity can be coherently described.[2]

This property — implementation independence — is not a limitation but a design principle. An OBP-conformant description of a sales process should be interpretable by any system that implements the protocol, regardless of whether that system runs on cloud infrastructure or local hardware, whether its agents are human or artificial, whether its storage is relational or document-based.

We call systems that implement OBP World Filters. Each World Filter translates the universal grammar of OBP into the specific vocabulary of its domain. A World Filter for a logistics company will instantiate OBP's primitives differently from one designed for a creative agency or a public sector institution. But because both are grounded in the same protocol, their descriptions remain mutually intelligible at the level of grammar.

This is the function of any genuine protocol: not to prescribe but to enable. TCP/IP does not specify what information should be transmitted; it specifies how transmission should occur such that any compliant system can participate. OBP does not specify what a business should be; it specifies the structure within which any business can describe itself coherently.

Three properties follow necessarily from this design.

Universality:  The grammar does not assume an industry, a technology, or an era. A trading house in 1850 and a network of autonomous agents in 2050 are both fully describable within the same four primitives — at different resolutions, but without loss of coherence.

Generative:  OBP is a grammar, not a list. It defines a set of primitives from which any valid description of business activity can be derived. The namespace does not enumerate concepts; it generates them.[8]

Implementation independence:  An OBP-conformant description is portable. It does not encode the assumptions of any particular technical platform. A reference implementation exists as proof of feasibility; it is not a constraint on conformance.

Authorship: The Configure Principle

One of the four bases — being — possesses a distinctive capacity that the others do not: it can configure itself and its environment.

In the vocabulary of OBP, this capacity is formalized as being.configure. It denotes the ability of any being to construct, modify, or extend the systems through which it operates. A business that can configure its own operational infrastructure — its workflows, its data structures, its agent capabilities — is not merely executing a fixed program. It is, in a meaningful sense, writing its own code.

This principle has significant implications for the design of business systems. A system that can only be configured by external specialists is dependent. A system that configures itself in response to its own operational experience is adaptive. A system in which the beings that use it can extend and modify its capabilities is, in the deepest sense, alive.

The configure principle is not a technical feature. It is a philosophical commitment: that the appropriate relationship between a being and the systems it inhabits is one of authorship, not mere consumption. Business systems should be written by the businesses that use them, in a language sufficiently powerful to express their full operational reality.

Without configure, business systems remain things that are configured by others. The configure principle inverts this: the system is the expression of the being's intent, not a constraint imposed upon it.

Sovereignty as a First Principle

OBP establishes data sovereignty as a structural invariant, not an optional feature.

In the language of OBP, the principal — the entity that owns and governs a given domain — is always defined. Every record, every process, every agent action can be traced to a principal who bears ultimate responsibility for it. This is not merely a matter of compliance or accountability, though it entails both. It is a claim about the nature of legitimate authority in any system that affects human lives.

The concept of portability follows from sovereignty. If a being owns its data — its records, its operational history, its accumulated knowledge — then that ownership must be portable. The data must be expressible in a form that is not dependent on any particular platform. A business that cannot move its own operational knowledge from one system to another is not, in any meaningful sense, the owner of that knowledge.

OBP's commitment to portability is not anti-competitive. It is constitutive of competition itself. Markets function when participants can freely choose among alternatives. A business trapped in a platform it cannot leave has no meaningful choice. OBP provides the grammatical foundation for a world in which exit is always possible.

Every entity that operates within an OBP-conformant system has a defined owner, a defined domain of authority, and a defined set of rights to its own operational data. These rights are not granted by the platform; they are guaranteed by the protocol.

The Fluctuation: Human in the Protocol

OBP makes a specific claim about the status of human beings within business systems that distinguishes it from most operational frameworks.

A human is not simply a user of a business system. Neither are they simply a resource to be allocated. In OBP, a human occupies two positions simultaneously.

As a principal, the human is the ultimate sovereign. The human authorizes agents, owns data, and bears ultimate responsibility for the operations conducted under their authority. No agent action is legitimate without traceable authorization from a principal.

As a being, the human participates in workflows, produces resources, and acts within contexts like any other operational entity. At this level, a human is structurally indistinguishable from an AI agent: both receive tasks, both produce outputs, both operate within the protocol's grammar.

This dual citizenship is not a contradiction. It reflects the genuine complexity of human agency in organizational contexts. We are simultaneously authors of our systems and actors within them, simultaneously governors and governed.[9]

The protocol acknowledges this complexity without resolving it. OBP does not prescribe the correct balance between human authority and agent autonomy. It provides the structural conditions under which that balance can be negotiated, adjusted, and renegotiated over time.

The tension between individual sovereignty and systemic order is not a problem to be solved. It is the engine of organizational life. A system that eliminates this tension — that fully automates the human role or that refuses all automation — has misunderstood the function of the protocol. OBP exists to maintain the space in which human beings can be genuinely human: unpredictable, creative, sovereign.

AI Governance and the Protocol's Vocabulary

OBP does not participate directly in policy debates. It is an engineering specification, not a position paper. But its design intersects structurally with concepts that recur across the academic and regulatory discourse on AI governance. The following correspondences are maintained as an index for researchers and policymakers who encounter OBP from that direction.

OBP Primitive Governance Concept Structural Role
being / Owner Sovereignty Sovereignty Structurally defines whose will an AI action is attributable to
flow.delegate Non-Delegation Principle Makes explicit and records the scope of authority delegated to agents
Agent Autonomy Gate Contestability Structural guarantee that a human can intervene when risk thresholds are exceeded
Trust Track Record Auditability Immutable record of all agent actions, available for inspection
context.guardian Accountability Explicit designation of the governance principal responsible for a domain
World Filter Jurisdictional Boundary Defines the scope — organizational, national, platform — within which OBP operates

OBP does not adopt a national or political stance. It functions as a neutral, open protocol. Its scope is the organizational layer: the grammar through which an entity's agents operate under that entity's sovereignty. Frontier model safety, national compute sovereignty, and international regulatory frameworks belong to other domains of responsibility.

OBP occupies the missing layer between high-level policy aspirations and low-level technical implementations. It does not replace governance discourse; it provides the structural vocabulary that governance discourse can refer to when asking: "Is this agent acting within the authority granted to it?"[10]

External Protocol Compatibility

OBP defines the semantic and governance layer of agent operations. It does not define how agents connect to tools, how they communicate with one another, or how data is transported over networks. These concerns belong to distinct layers, and OBP's relationship to them is one of composition, not competition.

Layer 3 — Semantics and Governance (OBP).  Who delegates what to whom. Owner Sovereignty definitions. The meaning of an agent's action within a business domain.

Layer 2 — Execution Infrastructure.  Tool connection protocols. Agent instruction formats. Inter-agent communication standards. OBP does not govern this layer.

Layer 1 — Transport.  HTTP, WebSocket, gRPC, and their successors. OBP does not govern this layer.

OBP governs Layer 3 exclusively. An OBP-conformant agent invokes tools through whichever Layer 2 protocol its execution environment provides. OBP's delegation primitives — flow.delegate, principal, actor class — function as higher-order vocabulary that sits above the execution layer, not within it.

Three principles govern OBP's relationship to external protocols:

Composition, not competition.  Layer 2 protocols are execution infrastructure that OBP-conformant systems utilize. They are not rivals to OBP; they are its foundation.

Sovereignty of specification.  OBP maintains an independent specification at the semantic layer. It is not absorbed into, nor subordinated to, any execution-layer standard.

Higher-order vocabulary.  OBP's primitives describe the meaning and authority of agent actions. Execution-layer protocols describe the mechanics. Both are necessary; neither subsumes the other.

Use Boundary

The OBP core specification is published under a permissive license. Any entity may read the specification, implement conformant agents, and declare conformance. Conformance testing will be formalized in a future version.

What is not permitted: intervention in OBP's specification governance process, redistribution of modified specifications under the OBP name, or claims of OBP conformance based solely on imitating a reference implementation rather than implementing the protocol itself.

The Butterfly Axiom

OBP makes a formal claim about the causal structure of the systems it describes.

Any single atomic change — one record created, one permission modified, one delegation authorized — can, given sufficient time and a system operating in a nonlinear regime, produce effects of arbitrary magnitude on the state of the entire domain. This is the Butterfly Axiom: not a metaphor, but a structural property of the systems OBP is designed to govern.

The axiom has a precise formulation. For any initial perturbation δ greater than zero, and any threshold ε greater than zero, there exists a time t* at which the magnitude of the system's deviation exceeds ε. In domains exhibiting positive Lyapunov exponents — and empirical evidence from biological and economic systems confirms that many do — small perturbations grow exponentially.

The immediate consequence is a refinement of the observer principle. No act of observation or governance is without causal impact on the system being observed. A protocol that monitors agent behavior is itself an agent within the system. Complete separation between governor and governed is a theoretical limit, not an achievable state.

This is the OBP analogue of Heisenberg's insight: observation disturbs the system. The protocol acknowledges this rather than denying it.

The governance implication is direct. Because every intervention — no matter how small — is causally significant, the design goal of OBP governance is not zero intervention but minimization and transparency of intervention. Every action taken within an OBP-conformant system is recorded, attributed, and auditable. The Butterfly Axiom does not counsel inaction; it demands that every action be taken with full awareness that its consequences may propagate beyond the horizon of immediate prediction.

Conclusion: Toward the Open Standard

OBP v1.0 is a beginning. We publish this document not as a finished artifact but as a stake in the ground: a claim that the problem of business complexity is, at its root, a grammatical problem, and that grammatical problems admit grammatical solutions.

A reference implementation of OBP is under active development. We expect to publish it when the code and the protocol are sufficiently aligned that reading one illuminates the other. Until then, this document stands as the authoritative statement of the protocol's principles.

The long-term ambition of OBP is to become a public good: a standard maintained by a community of practitioners, implementors, and researchers, owned by no single entity, controlled by no single interest. We are aware that this is a large ambition and that the distance between a first publication and a genuine public standard is measured in years of patient work.

We publish now because the problem is urgent. The proliferation of AI agents, each operating within its own proprietary vocabulary, threatens to reproduce at an accelerated pace the very fragmentation that business software has suffered for decades. The window for establishing a common grammar is narrowing.

The protocol is open. Its development will be made public as it matures. Its governance will evolve as its community grows. This document, published on 2026-04-01, is the first public statement of OBP. It establishes priority of conception and invites scrutiny. We expect to be corrected, extended, and improved. We expect the protocol to become the DNA of business — outlasting its authors.

References

  1.  Luhmann, N. (1995). Social Systems. Stanford University Press.
  2.  Teece, D. J. (2007). Explicating dynamic capabilities: the nature and microfoundations of (sustainable) enterprise performance. Strategic Management Journal, 28(13), 1319–1350.
  3.  Latour, B. (2005). Reassembling the Social: An Introduction to Actor-Network-Theory. Oxford University Press.
  4.  Watson, J. D., & Crick, F. H. C. (1953). Molecular structure of nucleic acids: a structure for deoxyribose nucleic acid. Nature, 171, 737–738. doi:10.1038/171737a0
  5.  Monod, J. (1971). Chance and Necessity: An Essay on the Natural Philosophy of Modern Biology. Knopf.
  6.  Pacioli, L. (1494). Summa de arithmetica, geometria, proportioni et proportionalità. Venice.
  7.  Taylor, F. W. (1911). The Principles of Scientific Management. Harper & Brothers.
  8.  Chomsky, N. (1957). Syntactic Structures. Mouton. — The argument that a finite grammar can generate an infinite set of valid expressions is the direct antecedent of OBP's generative design.
  9.  Arendt, H. (1958). The Human Condition. University of Chicago Press. — The distinction between labor, work, and action — and the dual position of the human as both author of systems and actor within them — informs OBP's treatment of the principal/being duality.
  10. Ashby, W. R. (1956). An Introduction to Cybernetics. Chapman & Hall. — The Law of Requisite Variety: a regulatory system must possess at least as much variety as the system it governs. OBP's four bases are proposed as the minimal variety sufficient to describe any business system.

V.  Agent Constitution

Agent Constitution

OBP Specification Committee — 2026-04-07

An agent exists to reach where the will of a single being cannot reach alone.

The Agent Constitution defines what an agent is, its relationship to its principal, and the invariant duties it must uphold — independent of any implementation, product, or organization.

VI.  Lexicon

Terms defined by this protocol.

humanic adjective — CC0

Of or relating to the embodiment of a Human Principal's sovereignty, context, and intent. Applied to agents that function as genuine extensions of their owner — not as independent actors, not as simulators of compliance.

The term fills a structural gap in the language. Human agent denotes a human being acting as an agent — the opposite of the intended meaning. No existing adjective describes the state of an artificial agent that reliably embodies the will of its principal. Humanic is that word.

humanic agent — an agent that embodies its owner's sovereignty

humanic action — an action grounded in the Human Principal's intent

humanic context — context inherited from the Human Principal

This term is released into the public domain under CC0. No entity may claim exclusive rights to the word humanic as defined here.