Security
Built to handle other people’s customers
We handle your customers and their data, so security, privacy and reliability are foundations — not features. Here’s what’s in place, and what’s coming.
Live today
Protecting your data now
Workspace isolation
Every workspace’s data is isolated at the database level. One customer can never see another’s conversations or contacts.
Role-based access
A ten-level permission model controls exactly what each team member can see and do.
Secure authentication
Passwords are hashed with a strong algorithm and sessions are signed. Changing or resetting a password, disabling an account or choosing “sign out everywhere” ends every web and mobile session within a minute.
Two-factor authentication
Authenticator-app (TOTP) codes with single-use recovery codes, available to every user and mandatory for Aniro platform staff.
Encrypted in transit
All connections use TLS. Data moving between your browser, Aniro and the messaging platforms is encrypted.
Encrypted channel credentials
Access tokens and secrets for your connected channels are stored encrypted at rest.
Audit logging
Sensitive actions — support access, member and role changes, configuration edits — are recorded.
Data export & deletion
Your data is yours. You can export it and request deletion.
Responsible AI
AI you can safely put in front of customers
The Intelligence Engine is built to be trustworthy by default — grounded, supervised, and always under human control.
Grounded answers
The AI answers from your approved knowledge and cites its source — it doesn’t invent policy or prices.
Human-approved learning
Nothing becomes trusted knowledge until a person reviews and approves it. The AI never silently rewrites what it knows.
Pre-send safety checks
An AI supervisor checks every reply before it sends — blocking unsupported claims and asserting irreversible actions.
Knows when to hand off
When it’s unsure, or a customer asks, the AI hands the conversation to a human with a full summary.
Gated actions
Money-moving or irreversible actions (refunds, bookings, charges) always require human approval.
You stay in control
You decide when the AI replies, when it pauses, and what it’s allowed to do.
On the roadmap
What we’re building next
We’re transparent about what isn’t here yet. These are planned, not available today.
- PlannedSSO (SAML) and SCIM provisioning
- PlannedPasskeys (WebAuthn) and workspace-enforced two-factor
- PlannedSOC 2 Type II attestation
- PlannedISO 27001 certification
- PlannedConfigurable data-retention policies
Have a security or compliance question? Get in touch — we’re happy to help.
Start answering every customer today
Connect your first channel in about two minutes. Every message lands here — answered and organised.