Skip to content
Operon AI
Security & Trust

What protects your calls and your data

Everything on this page is in the product today: encrypted connections and credentials, a workspace per company, a role per person, an audit log, and an agent that tells every caller it is an AI.

0.00%
voice path uptime
Trailing twelve months, measured at the carrier edge.
0
customer data breaches
Since the company was founded.
0h
vulnerability acknowledgement
72 hours to triage.
0
user roles
Owner, admin, operator, and viewer.
Controls

What is actually in place

Encryption in transit

Connections to the Operon console and API are encrypted with TLS. HSTS is on, so browsers refuse to fall back to an unencrypted connection.

Encrypted credentials

The keys and tokens that connect Operon to your other systems are encrypted with AES-256-GCM before they are stored in our database.

Workspace isolation

Each company works in its own workspace, and a workspace only ever sees its own data. Nothing one company stores is visible to another.

A role for every person

Everyone on your team signs in as themselves with one of four roles, owner, admin, operator, or viewer, so each person gets the access their job needs.

Audit log

Actions in your account are written to an audit log with who did it, what changed, and when, so a question about a change has an answer.

AI disclosure on every call

At the start of every call, the agent says it is an AI and tells the caller about recording. Nobody has to ask to find out.

Recording is your call

Your company can switch call recording off. Whether conversations are recorded is a decision you make, not one we make for you.

Claude, by Anthropic

The AI model behind every conversation is Claude, built by Anthropic. We name the provider so your reviewer does not have to ask.

Data handling

Your data stays inside your workspace

This is the question every operator asks, so here is the plain answer. Each company gets its own workspace, each person in it gets a role, and what they do is written down.

  • Console and API traffic encrypted with TLS, with HSTS on
  • Integration credentials encrypted with AES-256-GCM in the database
  • Each workspace sees its own data and nothing else
  • Owner, admin, operator, and viewer roles, set per person
  • An audit log of who changed what, and when
On every call
AI disclosureAt the start
Recording noticeAt the start
Call recordingCan be switched off
Language modelClaude, by Anthropic
Incident response
Vulnerability acknowledgement24 hours
Triage72 hours
Customer notificationWithin 24h of confirmation
Status pagePublic
On-call coverage24/7/365
Questions

What security teams ask us

If your reviewer needs something not covered here, we will answer it directly rather than pointing at a portal.

No. Operon has no SOC 2 report, does not sign HIPAA business associate agreements, and has no PCI DSS attestation. What is in place is what this page lists. If your business needs any of those, raise it before you sign rather than after.

Claude, the AI model built by Anthropic.

Yes. At the start of every call the agent says it is an AI and tells the caller about recording. Recording itself is your company's choice, and you can switch it off.

No. Each company works in its own workspace, and a workspace only ever sees its own data.

That follows the role you give each person: owner, admin, operator, or viewer. Actions in the account are written to an audit log, so you can always see who changed what and when.

Send it through the contact form with Security as the topic. We acknowledge within 24 hours and triage within 72.