Skip to main content

​
AI disclosure settings

The AI-disclosure settings surface holds your organization’s disclosure posture for every AI voice and chat agent: what the chat notice and voice intro say, which jurisdiction rules the workspace applies, and whether the disclosure is active at all. Configure it once under Settings → Compliance → AI Disclosure, or via the /api/v1/compliance/ai-disclosure endpoint described below, and every agent in the workspace reads the same row.
This is a tenant-owned control. You decide which jurisdiction regime applies to your traffic and what the notice says; the platform stores the setting and delivers it on every AI interaction while it is enabled. It does not mandate disclosure. Confirm which regimes apply to you with qualified counsel.

​
1. When a disclosure is required

Four regimes share the same disclosure shape — the contact must be told they are interacting with an AI — so the settings surface encodes all four with one toggle each: The jurisdictional rule toggles mark which regimes apply; they do not activate the notice by themselves. default_enabled is the master switch — it must be on for any notice to render or any AI-generated content marking to stamp. The EU AI Act page maps regulatory geography to the disclosure banner in detail, including the caveat below.
No automatic geo-detection. A rule toggle applies workspace-wide to every AI interaction; the platform does not infer a contact’s jurisdiction from their number, address, or IP. If your traffic touches the EU, enable the EU rule for the whole workspace. The one exception is the SB 243 minor reminder, which fires only for contacts you have explicitly flagged as a minor — minor status is never inferred from a birth date or any other signal.

​
2. How the settings endpoint works

Both calls require the owner or admin role. The workspace stores its disclosure posture in one singleton row, so a GET returns the full current state and a PUT accepts a partial update — only the keys you send change; everything else keeps its value. A fresh workspace returns { "settings": null } until it is first configured. Every write is recorded in the audit log with the acting user’s identity and the previous and new values of every changed field, so an auditor can see exactly what a mandatory AI-disclosure notice changed from and to. A PUT also takes effect immediately — the next agent turn reads the new values; you do not republish or redeploy agents when the posture changes. An invalid payload returns a 422 naming the field that failed.

​
3. How the banner text is composed

The text a contact sees comes from the same row: chat_notice_text for text channels and voice_intro_text (or your pre-recorded voice_intro_audio_url) for voice. The agent-identity governance surface — covered in Agent identity governance — is the inventory of every agent in your organization with its sponsor and channel reach: use it to confirm which agents exist and will carry the notice before you rely on it. Ready-made utterance wording per channel lives in Opt-in disclosure templates.

​
4. Activation-time vs runtime posture

Saving settings takes effect immediately — the very next agent turn reads the new values. But the pre-publication walkthrough is the right time to set this up: once you flip an agent from draft to active, it takes live traffic, and an unconfigured row means no notice renders at all. Before you publish any agent:
  1. GET /api/v1/compliance/ai-disclosure and confirm the row is not null.
  2. Confirm default_enabled is true and the jurisdiction toggles your traffic touches are on.
  3. Read the chat_notice_text and voice_intro_text exactly as a contact will see and hear them.
  4. Run a live check — start a chat session and confirm the notice appears on the first reply; place a test call and confirm the intro plays before the agent speaks.
At runtime the same endpoint is also your verification surface: audit every change under the audit action compliance.ai_disclosure_updated, and diff the row before a go-live review rather than assuming it still matches what legal approved. If a regime you must satisfy is not covered by the toggles above, contact trust@devotel.io.