Healthcare AI vendor risk · A ReMyll product
Sample assessment Contact
AITPRM Health™A ReMyll product
Trust Center

A review platform, not a clinical system.

Before you evaluate a vendor with us, your own teams will evaluate us. AITPRM Health assesses vendor security claims, architecture, subprocessor chains and policies. This page sets out what the platform does and does not hold, the controls in place, and what your security and privacy teams receive when they review us.

Trust and scope

What the platform processes, and what it does not.

The platform is not a PHI system. No field accepts protected health information, and no clinical data flows into it. We do not describe the platform as HIPAA compliant or SOC 2 certified, because we do not claim certifications we do not hold.

The platform processes
Vendor questionnaire responses
Vendor policy and architecture documents
Subprocessor registers and agreement status
Reviewer scores, ratings and rationale
Findings, severity tiers and sample contract controls
The platform does not process
Protected health information
Clinical records or encounter content
Patient identifiers of any kind
Live data flows from clinical systems
Model inference on patient data
Security

Controls in place, stated plainly.

Where a control is a configuration option rather than a default, this says so.

AccessSingle sign on and least privilegeSAML single sign on with SCIM provisioning, role based access control, and enforced multi factor authentication.
DataEncryption in transit and at restTLS 1.2 or above in transit, AES-256 at rest, with key rotation on a defined schedule.
IsolationTenant separationCustomer records are logically separated, and no customer data is used to improve the service for another customer.
ScopeNo PHI in the platformThe platform is designed not to collect protected health information, and customers are instructed not to enter it.
ResilienceBackup and recoveryEncrypted backups with defined recovery point and recovery time objectives, tested on a regular cycle.
AssuranceIndependent testingA documented vulnerability disclosure route and a defined remediation process for reported issues.
Framework mapping

NIST AI RMF 1.0, ISO/IEC 42001, HHS ASTP/ONC HTI-1 and CHAI.

Mapped and aligned describe how the platform is structured against these frameworks. They are not certifications.

Framework or functionHow the platform supports itReference
GovernDefined roles, a documented scoring rubric, published severity tiers and an audit ledger of every decision.Management system and roles
MapIntake and scoping establish deployment context, clinical setting, autonomy level and applicable statutory overlays.Context of the organization
MeasureWeighted domain scoring, evidence confidence rated separately, severity tiering with score ceilings.Performance evaluation
ManageFindings register, corrective action tracking, sample contract control generation and reassessment triggers.Operational planning and control
HHS ASTP/ONC HTI-1Predictive decision support intervention (DSI) transparency: source attributes for intended use, training data provenance, fairness measures and risk management disclosures are requested from the vendor in HTI-1 terms.DSI source attributes (§170.315(b)(11))
CHAI Assurance Standards GuideLocal validation testing before deployment, subgroup bias auditing across demographic and clinical cohorts, and continuous post-deployment monitoring with defined reassessment triggers.Assurance standards guide
Enterprise controls

What is default and what is contractual.

ControlAvailabilityNotes
Single sign on and SCIM provisioningEnterprise360Enforced at the tenant level
Role based access controlAll plansRoles driven by directory group membership where SSO is enabled
Data residency commitmentEnterprise360Stated in the agreement
Uptime commitmentEnterprise360Stated in the agreement
Full record exportAll plansAvailable at any time, including at exit
API accessRead only on ClinicalGuard, full on Enterprise360Custom and imported questions on Enterprise360

See plans and pricing →

Our subprocessors

We ask vendors to name their chain. Here is ours.

Every third party that processes data on our behalf, its function, its region and its agreement status. No subprocessor receives PHI, because the platform holds none. Updated on change; customers on Enterprise360 receive thirty days’ notice.

FunctionPurposeRegionAgreement
Cloud hostingCompute, storage, backupUnited StatesExecuted
Transactional emailNotifications to reviewers and vendorsUnited StatesExecuted
Authentication (SSO/SCIM)Identity federationUnited StatesExecuted
Error and uptime monitoringApplication telemetry, no customer contentUnited StatesExecuted
Document renderingExport of packets and ledgersUnited StatesExecuted
Provider names are disclosed to customers and prospects under NDA.One-page security overview →
Diligence

What your security team receives.

Documentation is available on request as the program matures.

QuestionnaireYour standard form, completedOn request.
ArchitectureData flow and hosting detailOn request. Where data is stored and processed, which regions are involved, and who can access it.
SubprocessorsOur own chain, enumeratedPublished. Every subprocessor we use, its function, its region and its agreement status.