How do you decide what a directive is worth?

From the action it proposes: the system of record it writes to, the value at risk, the role exercising authority, and the regulatory exposure. Enterprises tune the scoring; Orcher enforces it.

Where does the 40–85% range come from?

Observed spend reduction across deployments. Internal routing tests matched 91% of frontier-tier accuracy at 63% lower cost; that figure is Dipp AI research and is not independently verified.

Do reasoning models break the model?

They make it more necessary. Reasoning tokens multiply effective cost per call, which is why the tier decision has to be governed rather than defaulted.

Can a team be given its own budget?

Yes. Ceilings are set per role, department, and directive class, and they are enforced at call time rather than reconciled later.

Orcher · Component 06 of 07

Cost Governance

Routes by stakes and halts runaway loops.

Layer 2 — continuous, non-blocking

A routing manifold sorting requests by stakes, with a circuit breaker halting a runaway loop — Routes by stakes and halts runaway loops.

Not every directive deserves the frontier. Cost Governance routes by the stakes of the action rather than the habits of the developer, enforces budgets at the role and department level, and terminates loops that would otherwise spend without producing an outcome. Cost is a governance property, not a finance report written after the fact.

Selects the model tier from the consequence of the action, not the default.
Enforces budget ceilings per role, department, and directive class.
Detects and halts non-terminating agent loops.
Escalates to a higher tier only when verification demands it.

Without it, spend outruns value

Ungoverned agent fleets exhaust annual budgets in a quarter and produce no auditable outcome to show for it.

What it emits

A routing decision and a cost record per call, reconciled to the directive.

Evidence

What the record shows

Figures drawn from the Enterprise Superintelligence Report, Vol. I, August 2026.

40–85%

spend reduction observed under Orcher cost governance

range across deployments — Vol. I

4,500×

price spread across 400+ models and 70+ providers

Vol. I, p.25

63%

cost reduction at 91% accuracy match in internal routing tests

Dipp AI Research: internal analysis, not independently verified

68%

of enterprise AI programs are over budget

Vol. I, p.9

Mechanism

How it works

Four stages, in order. Layer 1 components gate execution; Layer 2 components run continuously and never block.

01
Score the stakes
Each directive is scored by consequence: what changes in a system of record, whose authority is exercised, and what a wrong outcome would cost. Stakes, not developer habit, set the tier.
02
Route to a tier
Across 400+ models and 70+ providers the price spread is roughly 4,500×. Orcher selects from that range deliberately, and escalates only where verification demands the stronger model.
03
Enforce ceilings
Budgets are enforced per role, per department, and per directive class at call time — not reported after the quarter closes.
04
Detect and halt loops
Non-terminating agent loops are detected and stopped, and the halt is written to the ledger like any other outcome rather than disappearing into a bill.

Operational contract

Input: Directive stakes score, role budget, and live provider pricing
Output: Routing decision and per-call cost record reconciled to the directive
Mode: Layer 2 — continuous, non-blocking
Controls: Tier selection, escalation policy, budget ceilings, loop termination
Attribution: Cost is attributed to a Verified Execution Cycle, not to an API key
Observed effect: 40–85% spend reduction across deployments

What it is not

It is not a FinOps dashboard

Dashboards describe spend that already happened. Cost Governance decides, at call time, whether the spend is permitted at all.

It is not a race to the cheapest model

High-stakes directives are routed to the strongest available tier on purpose. The discipline is matching the tier to the consequence, in both directions.

Failure behaviour

When a ceiling is reached or a loop is detected, execution halts and the directive is recorded as halted with its cost trace. Spend never continues silently past a limit.

Questions

What enterprises ask first

How do you decide what a directive is worth?: From the action it proposes: the system of record it writes to, the value at risk, the role exercising authority, and the regulatory exposure. Enterprises tune the scoring; Orcher enforces it.
Where does the 40–85% range come from?: Observed spend reduction across deployments. Internal routing tests matched 91% of frontier-tier accuracy at 63% lower cost; that figure is Dipp AI research and is not independently verified.
Do reasoning models break the model?: They make it more necessary. Reasoning tokens multiply effective cost per call, which is why the tier decision has to be governed rather than defaulted.
Can a team be given its own budget?: Yes. Ceilings are set per role, department, and directive class, and they are enforced at call time rather than reconciled later.

Where it shows up