Skip to content

AI you can put in front of an auditor.

Every AI action is logged, reviewable and subject to human approval, and your data never leaves the AWS boundary or trains anybody else’s model.

Powered by Amazon Bedrock

How it works

Amazon Bedrock, inside the AWS boundary.

Nothing leaves the boundary to be processed, and nothing you send becomes training data.

  • Human approval at every AI step
  • Full audit trail of every action
  • Your data is not training data
  • Encrypted in transit and at rest

Capabilities

The controls behind every AI action.

  • Human approval at every AI step

    Agents prepare work; people approve it. No AI step commits a decision on its own.

  • Full audit trail

    Every agent action and every human change is recorded against the case, field by field.

  • Your data is not training data

    Customer data is never used to train the underlying models, ours or anybody else’s.

  • Enterprise hosting

    SOC 2. HIPAA-aligned controls with BAAs signed. Third-party penetration testing and continuous monitoring.

Technical specifications

Under the hood.

What your IT, security and architecture teams will ask first.

Orchestration
BPMN-based, low-code and no-code process orchestration
AI models
Leading AI models via Amazon Bedrock, with the flexibility to choose the model that fits
Hosting
Amazon Web Services, multi-tenant, in Inspire's AWS account
APIs
API-first; the workbench UI runs on the same APIs customers use
Security & compliance
SOC 2; HIPAA-aligned controls with BAAs signed; third-party penetration testing and continuous monitoring
Integrations
Coherent Spark and the Healthmaps manual for rating; Gradient and Curv for claims data enrichment, AMS in build; Qlik for analytics

MSLX Solutions

MSLX solutions built on IPAS.

See MSLX at work in your process.

A demo built around your underwriting, claims, clinical or distribution team.

Book a Demo