Kurni Kwok AI shaping · operating-model design
Construct classification

AI shaping is a discipline, not merely a model, framework or method

AI shaping uses several related architecture terms, but they do not name the same kind of thing. This page separates the discipline, its design activity, the resulting operating model, its shaping approaches, durable mechanism, operational effect, capabilities and outcome.

Use this page whenThe question is what kind of construct AI shaping or one of its related terms is.
Choose another route whenThe question is how AI work develops, which tools execute it, or how AI shaping relates to broader governance and adoption frameworks.

Plain answer

AI shaping is the operating-model discipline. Operating-model design is a central design activity, the operating model is the resulting work-system arrangement, direct- and mediated-shaping approaches are alternative structural arrangements, shaped intelligence carrying a reusable and iteratively improvable domain work pattern and reusable work basis is the durable mechanism, two-sided work-burden shift is the operational effect, AI-shaping intelligence and shaped intelligence are capabilities, and productive AI work is the outcome.

Why the distinction matters

Calling every construct a model, method or framework obscures what is being applied, designed, carried or produced. The classification keeps public claims precise.

Current maturity

AI Shaping is in an early practice-derived codification stage

The seven AI Shaping principles are the highest strategic practice layer within the current body of work: established, practice-derived, evidence-sensitive and iteratively improvable under human direction. Fuller practitioner-facing specifications, architecture patterns, techniques, practices, anti-patterns and evaluation checks are maintained and further developed privately.

This does not make AI Shaping a formal standard, complete body of knowledge, universally validated method or implementation package.

How the public architecture uses each term

AI shaping is described as an operating-model discipline because it governs a coherent field of recurring practice: how broad AI capability is shaped into productive AI work through shaped intelligence carrying a reusable and iteratively improvable domain work pattern and reusable work basis, with source basis carried forward, two-sided work burden shifted and the human-authority boundary retained. Operating-model design is a central activity within that discipline, but the design activity is not the whole discipline. The operating model is the work-system arrangement being designed; AI-shaping intelligence and shaped intelligence are capabilities within that arrangement.

These labels are not universally mutually exclusive. A framework may express principles, a method may use a framework, and a capability may be established through an operating model. The distinctions below state how the terms are used in this public AI-shaping architecture.

TermWhat it means hereAI-shaping application
DisciplineA coherent field of practice with a defined purpose, concepts, governing conditions and recurring activities.AI shaping is the operating-model discipline.
PracticeThe actual application of a discipline in work.A person, team or organisation applies AI shaping under human direction.
Design activityThe intentional activity of arranging how something should operate.Operating-model design arranges the recurring human-and-AI work system.
Operating modelThe resulting arrangement of roles, capabilities, work, information, controls and authority.The work-system arrangement through which shaped intelligence carries recurring work under human direction.
ModelA context-dependent term that may mean an AI model, a conceptual representation or an operating model.Use the precise form: general-purpose AI model, conceptual model or operating model. Avoid using “AI-shaping model” as the category label.
Conceptual modelA representation used to explain relationships rather than to prescribe implementation.The concept stack and public diagrams explain how the discipline, approaches, mechanisms, capabilities and outcome relate.
ApproachA broad or structural way of addressing a problem or establishing a capability, usually less specified than a method.Direct-shaping approach and mediated-shaping approach are the two canonical structural alternatives. “Approach” remains too broad to classify AI shaping itself, which is the operating-model discipline.
FrameworkAn adaptable structure of concepts, components, relationships or rules.A framework may organise part of AI-shaping practice, but AI shaping is not defined merely as a framework.
MethodA more specified and repeatable way of performing work.Protected application may use specified methods and techniques, but the public category is broader than one prescribed method and the fuller practitioner material remains in development.
MethodologyA reasoned system or study of methods, though the word is often used loosely to mean a method.Not used as the canonical category label because its common meaning is ambiguous.
StandardAn agreed reference that states requirements, expectations or accepted guidance.This public release does not claim that AI shaping is a formal standard.
Guide or body of knowledgeAn organised reference that gathers and explains accepted knowledge or guidance.The website and papers explain the category, and a fuller practitioner-facing body is being codified privately, but no complete professional body of knowledge is claimed.
PrincipleA proposition that guides choices, conduct or evaluation.The seven AI Shaping principles form the highest strategic practice layer within the current body of work. They remain practice-derived, evidence-sensitive and iteratively improvable under human direction; they are not the formal category definition, a capability-evidence conclusion or implementation authority. Source basis and the human-authority boundary retain their separate classifications.
BoundaryA limit that must remain intact while work changes.The human-authority boundary prevents work-burden shift from becoming authority transfer.
Grounding conditionA basis that must remain connected to the work so the result is grounded.Source basis is the shared grounding condition around the two burden sides.
Durable mechanismThe reusable arrangement that enables the discipline to carry recurring work across changing instances.Shaped intelligence carries a reusable and iteratively improvable domain work pattern and reusable work basis.
Operational effectThe visible change in how recurring work burden is allocated.Two-sided work-burden shift is the operational effect and primary capability-evidence signal.
CapabilityA reusable ability to perform, shape or carry work.AI-shaping intelligence and shaped intelligence are capabilities with different roles.
Work systemInteracting people, sources, capabilities, controls and work states that jointly perform recurring work.The broader human-and-AI arrangement within which the discipline is applied.
ToolAn artefact used to perform particular tasks.A tool may support AI shaping, but it is not the discipline itself.
ProductA packaged offering supplied to users or customers.Neither AI shaping nor AI-shaping intelligence is defined merely as a product.
OutcomeThe intended result of applying the discipline.Productive AI work is the work outcome.

Compact architecture: AI shaping is the discipline; people, teams or organisations apply it; operating-model design is the central design activity; the operating model is the resulting work-system arrangement; direct- and mediated-shaping approaches are alternative structural arrangements; shaped intelligence carrying the reusable and iteratively improvable domain work pattern and reusable work basis is the durable mechanism; two-sided work-burden shift is the operational effect; source basis is the grounding condition; the human-authority boundary is the governing boundary; AI-shaping intelligence and shaped intelligence are capabilities; and productive AI work is the outcome.

A familiar project-management analogy

Project management provides a familiar example of why discipline, standard, guide, method, principles, framework, approach, operating model and capability should not be treated as synonyms.

ClassificationProject-management analogyAI-shaping application
Discipline or professional practiceProject management is the broad field, larger than any particular standard, guide, method, framework or lifecycle.AI shaping is the operating-model discipline.
Standard and guideThe Standard for Project Management and the PMBOK® Guide organise six core principles, seven performance domains and guidance for project-management practice.AI shaping is not presented as a formal standard or body-of-knowledge guide.
MethodPRINCE2® Project Management (Version 7) is a structured method containing seven principles, seven practices, seven processes, roles, people considerations and tailoring.AI shaping is broader than one implementation method. Methods may support its application without defining the whole discipline.
Values and principlesThe Manifesto for Agile Software Development establishes values and supporting principles that guide adaptive software development.AI Shaping establishes its seven principles as the highest strategic practice layer within the current body of work while retaining practice-derived, evidence-sensitive and non-standard maturity limits. The analogy concerns construct classification only and does not imply derivation, equivalence, institutional maturity or validation parity.
FrameworkScrum is a lightweight framework expressed through accountabilities, events, artefacts and commitments.No single public AI-shaping framework is claimed. A framework could organise aspects of practice without becoming the discipline itself.
Approach or lifecyclePredictive or Waterfall, adaptive or iterative, and hybrid lifecycles describe how project delivery proceeds.Organisations may adopt different approaches to applying AI shaping. Direct- and mediated-shaping approaches are structural alternatives, not equivalents of predictive and adaptive delivery.
Design activityDesigning project governance, roles, controls and reporting arranges how project work should operate.Operating-model design is the central design activity within AI shaping.
Operating modelAn organisation’s project-delivery operating model combines governance, decision rights, roles, information, processes, tools and controls.The AI-shaping operating model is the resulting human-and-AI work-system arrangement.
CapabilityOrganisational project-management capability is the reusable ability to initiate, govern and deliver projects effectively.AI-shaping intelligence and shaped intelligence are reusable capabilities with different roles.
OutcomeGoverned project delivery and intended benefits are results produced through the project system.Productive AI work is the intended work outcome.

Analogy limit: the comparison explains classification levels only. It does not claim that AI shaping is a project-management method, that direct- and mediated-shaping approaches correspond to predictive and adaptive delivery, or that AI-shaping intelligence corresponds to PMBOK, PRINCE2 or Scrum.

Public boundary

Public construct-classification explanation only: this page clarifies the public architecture. It is not an implementation method, formal standard, deployment assurance or method transfer.

Related pages

Reading sequence

Page 11 of 31