Security

Your supervision records, protected and private

Your staff, trainees and clients are real people, so we protect their records at every layer, down to each person's role.

  • Hosted on AWS
  • AES-256 encryption
  • Single sign-on

Platform capabilities

Built for agencies of every size.

From one clinic to many, the protections an IT review looks for.

  • Single sign-on

    Staff sign in through Microsoft Entra ID, Okta or any SAML 2.0 provider, on the Agency plan.

  • Multi-factor sign-in

    With single sign-on, your provider's multi-factor rules apply to every sign-in.

  • Access by role

    Admin, Supervisor, Supervisee and Viewer, or permissions you set one by one. The database enforces them.

  • Encrypted at rest

    AES-256 for the database and every file you upload, session videos included.

  • Encrypted in transit

    TLS 1.2 or higher on every connection, from your browser to the database.

  • Passwords we never see

    Sign-in runs on Amazon Cognito, so your password never reaches our servers.

  • Strong passwords

    At least 12 characters, with upper and lower case, a number and a symbol. Reset by email only.

  • AI assistants, no API keys

    Claude, ChatGPT and other AI assistants connect with OAuth and act as the signed-in person.

  • Deleted when you choose

    Delete sessions, notes and files yourself, any time, or ask us to delete all of your organization's data.

Infrastructure

An architecture that keeps records safe by design.

The walls between organizations are built into the database itself, not left to our code.

  • Runs only on AWS

    Serverless compute and a managed PostgreSQL database on Amazon Web Services, with no data center of our own.

    • AWS Lambda
    • Aurora PostgreSQL
    • Amazon Cognito
  • Each region on its own

    Canada and the United States are separate deployments, each with its own sign-in and database. Your records and files are stored in your region.

    • Canada app.ca.behaviewhq.com
    • United States app.us.behaviewhq.com
  • Walls in the database

    PostgreSQL row-level security limits every request to your organization. A missed filter in our code still cannot return another organization's records.

  • Tested on every change

    On every change, automated tests try to reach across organizations, and every attempt must be refused. A new table without row-level security fails the build.

    Refused across organizations

    • Read
    • Add
    • Change
    • Delete

How we run it

Discipline you can check.

Anyone can confirm the first three from a browser.

  • HTTPS only

    The app tells browsers to use HTTPS and nothing else, for two years.

  • Old encryption switched off

    Connections over TLS 1.0 and 1.1 are refused.

  • No framing

    No other website can show the app inside a frame, so a hidden page cannot trick you into a click.

  • Multi-factor for our staff

    Our staff use multi-factor sign-in for every production system.

Your data

What we keep, and never ask for.

Behaview asks only for what supervision needs.

What Behaview keeps

  • Names, emails, roles and teams
  • Skills, data and notes from each session
  • Training guides, comments and the files you upload

Each person sees only what their role allows.

What Behaview never asks for

  • Dates of birth
  • Government ID numbers
  • Diagnoses or medical records
  • Bank or salary details
  • Biometrics

Notes and custom fields are free text, so your team decides what goes in them.

Never sold. We do not sell or rent personal information.

AI

AI that works inside your permissions.

Behaview never uses your records to train an AI model.

  • Never used for training

    Behaview's own AI, such as Coach Me, runs Anthropic's Claude models on Amazon Bedrock, with permission only to run them. It cannot train a model on your records.

  • Sees only what you can see

    AI reads and writes through the same database walls as everything else, so it receives only what your organization could already read.

  • Your assistant acts as you

    Claude, ChatGPT and other AI assistants connect with OAuth and use your permissions. Turn off someone's role and their access ends at once.

Behaview's AI runs in the United States by default, for the strongest models. Processing in Canada is available on request.

Get in touch

Questions, reports and reviews.

  • Report a security issue

    Found a vulnerability? Tell us what you found, and we will follow up.

    security@behaviewhq.com
  • Privacy requests

    Ask what we hold about you, or ask us to correct or delete it.

    privacy@behaviewhq.com
  • Security reviews

    Send us your agency's questionnaire, or go through it with us on a call.

    Talk to Sales

Privacy Policy Terms of Service

Bring your security questions to a demo.

We will walk through your region, your roles and who sees what.