Trust & Security

Built for the enterprise.

WethosAI works with consequential enterprise decisions, behavioral models and sensitive organizational context. Security and privacy are built into every layer.

SOC 2 Type II

Attested

Annual independent audit of security controls, availability and confidentiality. SOC 2 is an attestation by an independent auditor, not a certification.

The report is available under NDA.

Security controls

Defense in depth for your most sensitive data.

  • Single Sign-On

    Sign in through your organization's identity provider. Access ends the moment your directory says it does.

  • Role-Based Access Control

    Granular permissions for who can run simulations, view results, and export data.

  • Data Isolation

    Each customer's data is logically isolated. Profiles and simulations are never shared across organizations.

  • Encryption at Rest & Transit

    AES-256 encryption for stored data. TLS 1.3 for all data in transit.

  • Audit Logging

    Comprehensive audit trails for all system access, simulation runs, and data exports. Exportable for your compliance team.

  • Penetration Testing

    Annual third-party penetration tests with results available under NDA for enterprise customers.

Privacy & data governance

Your data stays yours.

Customer data is isolated per organization and never used to train our foundation models. The two kinds of Twin are handled differently, because one involves the person and one does not.

How Twins are built

Permissioned Twins

A Permissioned Twin is built with that person's participation and consent, and they can see and withdraw it.

Observed Twins

People off the platform are modeled from customer-approved organizational context and relevant publicly available professional information. No assessment or account is required.

Responsible use

Responsible modeling. Human accountability.

WethosAI is designed to help organizations understand consequential decisions while respecting the people being modeled.

Purpose-defined modeling
Twins are developed using information relevant to authorized business and security use cases.
Responsible data use
Permissioned Twins involve direct participation and consent. Observed Twins use customer-authorized professional and organizational context.
Human review
Simulated decisions are modeled outcomes, not definitive evidence of how an individual will behave in real situations.
Workflow-focused findings
Security simulations are intended to help organizations evaluate decisions and strengthen controls, rather than grade employee performance.
Controlled access
Enterprise permissions help govern who can access models, conduct simulations and review results.

Data control

Control over where data lives and how long it stays.

Standard

No training on customer data.

Your simulations and Twins are never used to train our foundation models.

Data deletion on request.

Full data export and deletion within 30 days of request, with certification.

Regional data residency.

Choose where your data is stored: US, EU, or Asia-Pacific regions.

Optional enterprise configuration

Customer-managed keys.

Available for enterprise customers.

Zero-retention mode.

Enterprise customers can enable zero-retention mode where simulation data is processed but never persisted.

Security review

Have security or privacy questions?

Our team can support vendor assessments, security questionnaires and enterprise reviews. The SOC 2 Type II report and third-party penetration test results are available under NDA.