AI Disclosure & Voice Ethics

Dialo AI is built around one principle: every party on a call should know when an AI is on the other end of the line.

Our commitments

• Disclosure first. Every call placed through Dialo AI begins with a clear AI disclosure ("Hi, this is an AI assistant calling on behalf of…"). You cannot disable this.
• No voice cloning of real people without consent. Our voice catalog is limited to Cartesia-licensed voices. We do not allow uploads of recordings of real third parties for cloning.
• No undisclosed political or campaign calling. Dialo AI may not be used for AI-generated political robocalls in jurisdictions that prohibit them.
• No fraud or impersonation. Impersonating a specific human, government agency, or organization is prohibited and grounds for account termination.
• Recording requires consent. Where call recording is enabled, the AI explicitly asks the recipient for consent and abides by their answer.

Models we use

• • Speech synthesis: Cartesia (licensed voice catalog).
• • Conversation reasoning: OpenAI / Google models, configurable per script.
• • Post-call summaries: Google Gemini.
• • Telephony: Vapi (PSTN routing, on Business plan).

Reporting abuse

If you've received an abusive or non-consensual call from a Dialo AI agent, email abuse@dialo.ai with the date, time, and (if known) caller number. We investigate every report.