Trust you can hold us to.

We're building the layer that makes AI voice accountable. That obligation applies to us first — here's what we commit to.

Transparency by default

People have a right to know when they're talking to a machine. Disclosure is on by default and can't be silently switched off.

Consent is meaningful

Consent must be informed, specific, and revocable — captured in the caller's own words and honored immediately.

A human is always reachable

Automation should never trap someone. A path to a person is part of the protocol, not a setting.

Evidence over assertion

We don't ask to be trusted. Every disclosure and consent is recorded and independently verifiable.

Minimal data

We retain the least we can. No plaintext call audio by default; scoped access; residency controls.

Open to scrutiny

The protocol is public and the reasoning is published. Trust infrastructure has to be inspectable.

Accountability

If we ever fall short of these, tell us.

We publish a responsible-disclosure path and treat these commitments as testable, not decorative. Hold us to the standard we're asking the industry to adopt.