AUTHORITY CAN BE PROVISIONAL. BOUNDED. REVOCABLE.
Runtime is not a synonym for real-time. A regulated decision can be allowed now, within an explicit cap and clock, while slower controls continue — then confirmed, watched, restricted or revoked when new evidence arrives.
A PROVISIONAL PASS IS A
FIRST-CLASS DECISION,
NOT A MISSING CHECK.
Live Case 01 asked whether evidence can earn authority. Live Case 02 asks whether authority can be explicitly bounded, time-limited and later changed when deferred controls resolve.
Two clocks, one case
FAST ACTION · SLOW ASSURANCEThe scalable pattern is not “run every expensive check before every action”. It is to separate blocking controls from deferred work and make the temporary authority explicit.
Blocking checks
Cheap / essential controls finish first. A hard red can stop the case immediately.
Provisional authority
Blocking pass + deferrable pending → limited authority with cap + TTL.
Deferred evidence
EDD, external vendors, human review and other slow controls continue asynchronously.
State transition
New evidence moves the case to CONFIRMED, WATCH or RESTRICTED. Silence cannot extend provisional status forever.
A decision is a trajectory
STATE MACHINEThe system does not store one boolean “approved”. It stores the current state, the reason for it, the clock, the limits and the event that caused each transition.
PENDING_T0 ├── hard red ─────────────────────────→ BLOCKED_T0 │ └── blocking controls complete + hard red absent + deferred control pending ↓ PROVISIONAL ├── deferred clear ─────→ CONFIRMED ├── yellow/ambiguity ───→ WATCH ├── late fail/red ──────→ RESTRICTED └── TTL expires ────────→ WATCH / RESTRICTED WATCH ├── concern resolved ─────────────────→ CONFIRMED └── fail / timeout ───────────────────→ RESTRICTED
Control taxonomy — pending is not pass
ROLES, NOT VIBESControls need explicit roles. A missing result must never be silently interpreted as a successful check.
▚ BLOCKING
Must finish before authority.
Example: a hard-list hit or other admission condition the institution refuses to defer.
▚ DEFERRED
May finish after bounded admission.
Allowed only when the policy explicitly specifies the pending control, cap, TTL and review behavior.
▚ ADVISORY
Informative, not automatically blocking.
Signals that can influence monitoring or review without silently acquiring veto power.
Facts + FOL — make assumptions explicit
PROLOG SHAPE · NOT LAWFOL does not decide what the law “really means”. It reasons over a versioned institutional assumption set selected by the institution.
case(case_001).
policy_version(case_001, policy_2026_09_v1).
assumption_set(case_001, assumptions_2026_09_v1).
control_type(sanctions, blocking).
control_status(case_001, sanctions, pass).
control_type(document_integrity, blocking).
control_status(case_001, document_integrity, pass).
control_type(source_of_funds, deferred).
control_status(case_001, source_of_funds, pending).
blocking_controls_complete(case_001).
hard_red_absent(case_001).
admission_risk_within_threshold(case_001).
deferred_control_pending(case_001).
provisional_cap(case_001, amount_gbp, 500).
provisional_cap(case_001, transaction_count, 3).
provisional_ttl_hours(case_001, 24).
ai_authorised_to_override(case_001, false).
decision(Case, provisional_allow) :- blocking_controls_complete(Case), hard_red_absent(Case), admission_risk_within_threshold(Case), deferred_control_pending(Case). decision(Case, confirm) :- all_required_controls_complete(Case), hard_red_absent(Case), no_open_yellow(Case). decision(Case, watch) :- control_status(Case, _, yellow), hard_red_absent(Case). decision(Case, restrict) :- control_status(Case, _, fail). decision(Case, review_required) :- current_state(Case, provisional), provisional_expired(Case), deferred_control_pending(Case).
Versioned assumptions — interpretation becomes inspectable policy
PROVENANCE > NUMBERSThe scarce object is not a model score. It is the documented choice: what may be deferred, for how long, under which limits, and what happens when the clock expires. Each choice records where it came from — an external instrument the institution is bound by, or an internal discretion it has taken and can be asked to defend.
assumption_set: id: ASSUMPTIONS_2026_09_V1 supersedes: ASSUMPTIONS_2026_06_V2 approved_by: <named officer> approved_at: 2026-09-01 sanctions: may_be_deferred: false provenance: basis: external # bound by an instrument, not chosen source: "<instrument reference>" source_of_funds: may_be_deferred: true maximum_delay_hours: 24 provenance: basis: internal # discretion taken, defensible on request source: "risk appetite, approved 2026-09-01" provisional_admission: maximum_amount_gbp: 500 maximum_transactions: 3 aggregate_exposure_gbp: 500 # per counterparty, across concurrent cases provenance: basis: internal recurring_review: enabled: true interval_hours: 168 provenance: basis: internal unresolved_yellow: action: WATCH expired_deferred_control: action: ESCALATE_TO_HUMAN
THIS IS NOT A CLAIM THAT A LAW REQUIRES THESE VALUES — IT IS AN EXAMPLE OF HOW AN INSTITUTION COULD ENCODE AND VERSION ITS OWN OPERATING INTERPRETATION. THE PROVENANCE FIELD MATTERS MORE THAN THE NUMBERS: “WE DEFER THIS BECAUSE THE INSTRUMENT PERMITS IT” AND “WE DEFER THIS BECAUSE WE DECIDED TO” ARE DIFFERENT POSITIONS TO DEFEND, AND A SUPERVISOR WILL ASK WHICH ONE APPLIES. A SET WITH NO EXTERNAL BASIS ANYWHERE IS NOT AUTOMATICALLY WRONG — IT IS SIMPLY ENTIRELY THE INSTITUTION'S OWN RISK APPETITE, AND SHOULD BE LEGIBLE AS SUCH.
What may happen now?
FOL DERIVES · POLICY BOUNDSFOL derives the formal state. Policy-as-code turns that state into a bounded operational action.
{
"action": "ALLOW_LIMITED",
"cap_gbp": 500,
"ttl_hours": 24,
"pending": ["source_of_funds"],
"ai_override": false
}{
"action": "ALLOW",
"limits": null,
"ai_override": false
}{
"action": "ALLOW_LIMITED",
"review_required": true,
"cap_gbp": 250,
"ai_override": false
}{
"action": "RESTRICT",
"escalate_to_human": true,
"ai_override": false
}Evidence packets — a lifecycle, not one report
APPEND-ONLY IN SPIRITEach meaningful transition creates a new packet linked to the previous one. The evidence record becomes append-only in spirit, rather than a single mutable approval flag.
T0 decision
Blocking controls, assumption set, pending controls, cap and TTL.
Provisional state
Operational authority granted within explicit limits.
Counterparty notice
The only outward-facing packet. States that authority is provisional, under which limits, and when a closing determination is due.
Deferred result
New evidence + previous state + transition trigger.
Closing state
Final controls, final action and hash of the previous packet.
Policy revision
Emitted only when a case outcome changes the rule. Links the incident to the assumption set it superseded.
PACKET 003 EXISTS BECAUSE A PROVISIONAL PASS THAT THE COUNTERPARTY IS NOT TOLD ABOUT IS NOT PROVISIONAL — IT IS AN UNDISCLOSED RESTRICTION. EVERYTHING ELSE IN THE CHAIN LOOKS INWARD, AT AUDIT AND SUPERVISION. THIS ONE LOOKS OUTWARD.
{
"schema": "sdp-transition-packet-v1",
"case_id": "case_001",
"sequence": 3,
"previous_packet_sha256": "...",
"previous_state": "PROVISIONAL",
"new_state": "CONFIRMED",
"trigger": { "type": "CONTROL_RESULT", "control": "source_of_funds", "result": "PASS" },
"policy_version": "policy_2026_09_v1",
"assumption_set": "assumptions_2026_09_v1",
"formal_decision": "confirm",
"institutional_action": "ALLOW",
"ai_authorised_to_override": false
}After the error — a control plane that cannot learn is not a control plane
REVISION AS MECHANISMNobody is asked for a system that never fails. A long run with no internal error reports is itself a finding. What is asked for is that failure has a defined path back into the rules.
Know the general rules
The obligations that apply, separated from the reading you have chosen.
State your own risk
Given this product and counterparty base: which risks you accept, which you refuse.
Encode it
That judgement as controls, limits and a versioned assumption set — not prose.
Report it
An auditable record per transition, with the interpretation in force at the time.
Say what happens when it breaks
The remediation path, and the mechanism by which an incident revises the rule. The fifth is the one most specifications omit.
{
"schema": "sdp-policy-revision-packet-v1",
"sequence": 6,
"previous_packet_sha256": "...",
"trigger": { "type": "INCIDENT_REVIEW", "case_id": "case_001",
"finding": "deferred control resolved FAIL after provisional exposure" },
"assumption_set_before": "assumptions_2026_09_v1",
"assumption_set_after": "assumptions_2026_10_v1",
"changed": [{
"key": "source_of_funds.maximum_delay_hours",
"from": 24, "to": 4,
"reason": "observed loss window exceeded appetite"
}],
"approved_by": "<named officer>",
"ai_authorised_to_override": false
}Assumption sets are never edited in place. A rule change is an event with a cause, an author and a hash — so the question “why did you allow that, and what did you change afterwards?” has an answer that can be checked rather than recalled.
Hard invariants — things the model cannot negotiate
THE IMPORTANT PROPERTIES BECOME TESTS, NOT PROSEReversibility gate — where this page ends and LC03 begins
CROSS-CASE INVARIANTPROVISIONALITY EXISTS ONLY WHERE THE ACTION IS REVERSIBLE.
Everything on this page — cap, TTL, monotonic narrowing, deferred controls — is licensed by one fact: the actions it governs can be undone, clawed back or bounded after the fact. A payment can be recalled within the scheme's windows; a mandate can be revoked; a corridor can be narrowed to zero.
Disclosure is not reversible. Once identity data, ID documents or transaction history leave the institution, there is no cap and no TTL that can retrieve them. So for any irreversible outbound action, the provisional machinery on this page does not apply at all: the case may remain provisional internally, but the outbound packet may not.
Scaling hypothesis — reasoning on state changes, not every transaction
DESIGN HYPOTHESIS · NOT A CLAIM ABOUT ANY NAMED INSTITUTIONASYNC CONTROL PLANE: EXPENSIVE REASONING WHEN FACTS CHANGE. THIS PAGE DOES NOT CLAIM THAT ANY NAMED INSTITUTION CURRENTLY IMPLEMENTS THIS ARCHITECTURE — IT IS A DESIGN HYPOTHESIS FOR HOW THE SAME AUTHORITY-SEPARATION IDEA COULD BE DECOMPOSED FOR SCALE.
Minimum reproducible trajectories — four paths are enough
NEXT BUILD · DELIBERATELY SMALLThe next build should stay deliberately small. Four deterministic trajectories can prove the time-bounded authority concept without pretending to be a bank.
Transfer — the same machine, with an agent as the subject
ONBOARDING → MANDATEThe state machine does not depend on the counterparty being a person. The banking case has decades of operational evidence behind it, which is why it is used to work the mechanics out. The machine-actor case is where the same mechanics have no incumbent answer.
▚ CASE → MANDATE
The subject is an agent acting for a principal, not a customer being admitted.
▚ CAP → SCOPE
Bounded exposure becomes bounded permission: which actions, which systems, which counterparties.
▚ TTL → VALIDITY WINDOW
The mandate expires on the clock rather than persisting until someone revokes it.
▚ DEFERRED → PENDING ATTESTATION
Human confirmation, an external check, or a credential status that has not yet resolved.
▚ RESTRICTED → REVOCATION
Late evidence narrows or withdraws the mandate — and the receiving side can see that it did.
▚ PACKET CHAIN → ATTRIBUTABLE TRACE
On whose authority the agent acted, under what interpretation, and what changed afterwards.
A provisional mandate is how an agent can be allowed to act before every attestation has resolved — without the receiving organisation having to take the sending organisation's word for it.STATED PLAINLY · LIVE CASE 02
Definition of done
WHEN LC02 BECOMES “EXECUTED”, NOT “SPECIFIED”A · PROVISIONAL → CONFIRMED B · PROVISIONAL → WATCH C · PROVISIONAL → RESTRICTED D · PROVISIONAL → TTL_EXPIRED → ESCALATE
CI ASSERTS
- blocking controls complete before provisional authority
- caps preserved · TTL enforced
- late fail changes authority · AI override false
- packet hash chain valid · policy_version preserved · assumption_set preserved
- caps never widen while a control is pending · aggregate exposure ceiling holds across concurrent cases
- expired cached state refuses to authorise · counterparty notice emitted on provisional grant
- revision packet supersedes, never edits
UNTIL THOSE TRAJECTORIES EXIST AND REPRODUCE IN CI, THIS PAGE REMAINS A SPECIFICATION / BUILD-NEXT NOTE.
Summary — the line so far
LC01 → LC02 → LC03LC01 · DEMONSTRATED
Evidence must earn authority. The model explains — it cannot overrule a failed deterministic control.
open LC01 ↗LC02 · SPECIFIED
Authority can be provisional, bounded and revocable — cap + TTL + monotonic narrowing, with a receipt per transition.
you are hereLC03 · NEXT
Disclosure is irreversible — so provisional machinery stops at the boundary. HOLD = zero outbound PII.
open LC03 ↗