Enterprise use cases
Runtime authorization across regulated workflows
The pattern is constant — evaluate, decide, issue the visa, record. We start where an agent's action has the clearest consequences: financial services. The policies, authorized reviewers, and thresholds adapt to each industry that follows.
Regulated finance
Financial services first
Financial services
Agents initiate payments and transfers, change accounts and limits, work KYC/AML exceptions and remediation, adjudicate claims and send regulated customer communications. Crelis evaluates each consequential action against your policy before it executes and routes what needs a person to an authorized reviewer.
- Payments and transfers
- Account and limit changes
- KYC/AML exceptions and remediation
- Claims
- Regulated customer communications
Our design partnership is built to prove safer production deployment, faster approval of agent workflows, and less manual review and audit effort — on one bounded workflow, which can contain several execution tasks.
Additional regulated use cases
The same authorization loop, wherever an agent's action has consequences
Customer operations, healthcare, legal, IT operations and public sector follow the financial-services pattern: policy evaluates each consequential action, an authorized reviewer decides when policy calls for it, and each decision leaves a record.
Regulated workflow
Customer operations
Support agents resolve tickets autonomously at volume. Refunds, account changes, and edge cases above policy limits require human approval from an operator without breaking the queue.
- Policy-bound resolution within limits
- Seamless human handoff
- Per-ticket execution log
Regulated workflow
Healthcare operations
Clinical and admin agents touch protected data. Changes that policy marks as consequential require human approval from an authorized clinical reviewer, with each access and approval sealed in a tamper-evident record.
- HIPAA-aware routing policies
- Authorized clinical reviewer pools
- Tamper-evident access records
Regulated workflow
Legal & procurement
AI redlines contracts and drafts purchase orders at machine speed. Policy decides which clauses and amounts require human approval from an analyst, so nothing binding ships unreviewed.
- Clause- and amount-level policies
- Analyst approval queues
- Signed decision history
Regulated workflow
IT operations
Ops agents patch, provision, and clean up. Destructive or fleet-wide actions require human approval from an operator before they execute, and each change is recorded against the resource it touched.
- Blast-radius policies
- Operator approval gates
- Change audit per resource
Regulated workflow
Public sector
Agencies adopt AI under strict accountability mandates. Crelis provides the decision evidence: who approved what, when, under which policy — exportable for oversight.
- Policy-mapped approvals
- Oversight-ready exports
- Citizen-data safeguards
Scenes from the Crelis launch film.
The common thread
Evidence, not assurances
Whatever the industry, the output is the same: a sealed record proving what AI did, what humans approved, and which policy governed it. That's what auditors, regulators, and boards actually ask for.
Audit event stream · illustrative
Illustrative datasealed
14:02:11Z · Finance Agent
Proposed wire transfer $250,000 → ACME-7741
14:02:11Z · Crelis
Inputs evaluated · policy WIRE-HIGH fired · REQUIRE HUMAN APPROVAL · risk context 91
14:02:12Z · Crelis
Routed to treasury reviewer pool (2 eligible) · visa withheld
14:06:48Z · Treasury reviewer
Checked counterparty, approved with note
14:06:49Z · Crelis
Execution visa issued · Ed25519-signed, single-use
14:06:49Z · Banking endpoint
Visa verified · wire released — ref TXN-88231
14:06:49Z · Crelis
Evidence recorded · hash 0x91f3…aa07