Security
Effective May 5, 2026
Security is foundational to the Medicare Intelligence OS. We process compliance-critical data — call content, commission records, beneficiary interactions, and carrier intelligence — and we operate under HIPAA-aware infrastructure with controls aligned to industry frameworks. This page summarizes our current posture; customer-facing audit reports and questionnaires are available under NDA.
Notice
This page is informational and reflects current practice. It is not legal advice. For binding terms applicable to a specific engagement, refer to your signed agreement with Artificial Bridge LLC. Material changes will be posted here with a new effective date.
01Architecture and data protection
- ·TLS 1.2+ enforced for all client and server-to-server communication.
- ·AES-256 (or equivalent) encryption at rest for primary stores, backups, and object storage.
- ·Tenant isolation by logical separation in shared infrastructure, with carrier and FMO tenants supported by dedicated logical environments where required.
- ·Secrets management via a hardened vault with rotation and audit.
- ·Production data is not used in non-production environments.
02Access controls
- ·Role-based access control (RBAC) with least-privilege defaults.
- ·SSO and SAML 2.0 supported for FMO and carrier tiers; SCIM provisioning available.
- ·MFA required for all employee access to production systems.
- ·Just-in-time elevation and audit logging for privileged actions.
- ·Quarterly access reviews and automatic revocation on role change or termination.
03Application security
- ·Secure SDLC including code review, dependency scanning, and SAST in CI.
- ·Pre-deploy automated tests for authentication, authorization, and tenancy.
- ·Customer-data redaction for AI prompts where feasible; PHI is processed under the applicable BAA.
- ·Rate limiting, anomaly detection, and abuse mitigation at the edge.
04Vulnerability management
- ·Continuous dependency monitoring and patch SLAs aligned to severity.
- ·Annual third-party penetration testing; report summaries available under NDA.
- ·Internal red-team exercises against compliance and AI-misuse scenarios.
- ·Coordinated vulnerability disclosure: report findings to security@theartificialbridge.com — we will acknowledge within two business days.
05Monitoring and incident response
- ·Centralized logging with immutable retention for security-relevant events.
- ·SIEM-style detection on authentication, authorization, and data-access anomalies.
- ·Documented incident-response plan with on-call rotation and severity tiers.
- ·Customer notification timelines defined contractually; HIPAA Breach notifications consistent with 45 CFR § 164.410.
06Business continuity
- ·Geographically redundant infrastructure for primary services.
- ·Documented backup strategy with regular restoration testing.
- ·RTO and RPO targets defined per service; details available to enterprise customers on request.
07Compliance posture and roadmap
- ·HIPAA-aware infrastructure and BAA available for carrier and FMO tiers.
- ·SOC 2 Type II — engagement in progress; report distribution under NDA on completion.
- ·Alignment to NIST 800-53 (moderate baseline) and NIST AI RMF for AI-specific risk.
- ·CMS Digital Health Technical Assistance and ONC/FHIR alignment on the BRIDGEt App Library pathway.
08Procurement and due-diligence kit
Carrier and FMO procurement teams can request a due-diligence package under NDA. The kit is assembled from the same evidence the platform generates continuously, so it reflects current posture rather than a point-in-time snapshot.
- ·SOC 2 report — Type I and bridge letter now; Type II on completion of the engagement in progress.
- ·HIPAA security risk assessment summary and BAA template v2.
- ·Penetration test executive summary with current remediation status.
- ·Architecture and data-flow diagrams annotated with PHI trust boundaries.
- ·Hash-chained audit log sample mapped to SOC 2 CC4.1 and CC7.2 controls.
- ·Subprocessor list and completed CAIQ / SIG-Lite security questionnaire.
09AI safety and integrity
- ·Model outputs are gated by guardrails for CTM, AHIP, and CMS Marketing Guideline categories where applicable.
- ·Training data is sourced under permitted-use rights; PHI is segmented and processed only under BAA terms.
- ·Outputs are logged with model and prompt metadata to support audit and contestability.
- ·Customers may opt out of using their content for model improvement at the tenant level.
10Subprocessors
We use a limited set of vetted subprocessors for cloud infrastructure, model serving, observability, communications, and payments. A current list and notification of material changes are available to customers on request and as required by data-protection terms.
11Reporting a security issue
Please email security@theartificialbridge.com with reproduction steps, impact, and any supporting artifacts. We accept reports in good faith and will not pursue action against researchers acting in compliance with this disclosure policy.
Security & vulnerability disclosure
security@theartificialbridge.comArtificial Bridge LLC · Anna, Texas