A term is owned by the chapter that first defines it, and a later chapter may use it freely but may not redefine it. These definitions are the ones the continuity ledger records, which is the same record the continuity gate checks every chapter against.

Access price Chapter 1

The stated amount an organization pays for the right to use an AI capability, usually per seat or subscription. It does not include additional costs based on how much AI the organization actually uses.

AI Business Economics Chapter 3

The science of the business economics of consuming AI: an ordered body of conditional claims, each stating when it applies, what it concludes, and which other claims it depends on.

AI Operations Management Chapter 3

The management discipline that acts on AI Business Economics: the practice by which an organization buys AI capacity, plans its use, records and assigns consumption, allocates capacity, and holds that consumption to account.

Allocation and routing Chapter 3

The function that uses a stated rule to decide what runs first when capacity is limited and to send each piece of work to suitable capacity.

Category error Chapter 1

The error of managing a metered resource as if it were licensed software. It rests on a false assumption: that the access price fixes the total cost. The tool is the right one and the vendor is not at fault. The economic model applied to the tool is wrong.

Consumption event Chapter 1

A single use of an AI system that consumes metered computing resources and creates an underlying cost greater than zero, whether or not the buyer is charged for it separately. A consumption event may be measured in tokens, requests, credits, compute time, or another equivalent unit. It is the basic unit of consumption tracked in AI Operations Management.

Cost-and-value flow Chapter 2

The stream of money spent on a deployment together with the business return attributed to that spending. The two halves form one flow because neither can support an economic decision on its own.

Cost-value asymmetry Chapter 2

The condition in which metering records a deployment’s cost automatically, while its value is measured only when the organization has deliberately built a way to measure it. As a result, the organization may know precisely what the deployment costs without knowing what that spending produces.

Flat-rate objection Chapter 1

The claim that a flat per-seat price makes AI equivalent to licensed software; answered by meter relocation.

Flow Chapter 2

A continuous stream of related activity within a deployment, described by its rate, direction, and the records kept about it. An organization manages a flow by governing it over time rather than approving the deployment once.

Founding Questions Chapter 3

The five questions AI Operations Management exists to answer from an organization’s own records, one for each of its five functions.

Function audit Chapter 3

The six-step method for finding which of the five functions an organization performs for a deployment: list its AI activities; assign each to a function or a neighboring discipline; mark each function as performed, partly performed, or not shown; name who performs each; and state the first gap to close.

Holding value to account Chapter 3

The function that judges whether a deployment pays for itself within a value boundary set before spending begins and names one person responsible for knowing the answer.

Meter relocation Chapter 1

The placement of consumption metering on the provider’s side of a flat-rate subscription. The buyer pays a fixed price for access, while the provider continues to measure actual use. The provider sets the flat rate based on expected consumption and may change the price, allowance, or usage limits when actual consumption exceeds that expectation.

Metered resource Chapter 1

A resource whose consumption is measured per unit of use, here in tokens or their equivalents.

Metering and attribution Chapter 3

The function that records what was consumed and assigns each unit of consumption to the team, product, or purpose it served.

Planning and budgeting Chapter 3

The function that states how much AI an organization expects to consume in a period and detects when actual consumption departs from that expectation.

Record flow Chapter 2

The stream of information an organization keeps about its own usage, including which team or product consumed what, at what volume, and when. The provider always keeps a record because it must produce a bill. The buyer has a record only if it deliberately builds one.

Resource consumption model Chapter 1

The economic model in which deployed AI is consumed as a metered resource. Each use creates a consumption event, and cost accrues with each event. It contrasts with the software access model.

Software access model Chapter 1

A purchasing model in which an organization pays a fixed price for access to software, usually per seat or subscription. Because additional use creates little or no additional cost, the organization manages who has access rather than how much of the software each user consumes.

Sourcing Chapter 3

The function that matches the AI capacity an organization buys to the requirements of its work: which models and providers serve which work, and on what terms.

Three-flow mapping Chapter 2

The diagnostic that establishes the condition of the usage, record, and cost-and-value flows in a named deployment. The mapping also identifies the evidence that would change each diagnosis.

Usage flow Chapter 2

The stream of consumption events a deployment generates, including requests made, work performed, and outputs returned. It is the only one of the three flows that an organization normally seeks to increase.

Value boundary Chapter 3

The boundary within which an organization judges whether a deployment pays for itself, defined by the costs and value it counts, the period it covers, and the outcomes against which it measures performance.