Skip to content

Launch

How it works under the hood, in brief

You don’t need to understand the technical details to use the channel, but it’s worth knowing what it actually does for you:

  • Checks the mailbox every minute. The agent checks for new emails at the frequency you set in “Sync interval” (section 4) — 60 seconds by default, with a minimum of 30. There’s no need to forward anything or turn on anything else: as soon as a new email arrives at the mailbox, the next sweep picks it up.
  • Replies in the same thread. The agent’s reply arrives as a normal email reply — same subject with “Re:”, and the thread stays linked in your usual mail client (Gmail, Outlook…) just as if a person had replied.
  • Every automatic reply carries a transparency notice. By law (AI Act, Art. 50), every reply the agent generates includes a fixed notice at the end, which can’t be turned off from any setting: “This reply was generated by [your business]‘s AI assistant. Reply to this email to continue the conversation; a member of our team can step in at any time.”
  • It never replies to an auto-reply. If the incoming email is itself an automatic reply — an “out of office”, a vacation notice, or any email marked by its origin as automatically generated — the agent detects it and doesn’t reply. It’s a deliberate safeguard: without it, two automated systems (yours and the sender’s) could end up replying to each other endlessly.

Note — Legal basis at a glance

The email transparency notice complies with the same AI Act Art. 50 you already know from the widget (Manual 24, section 6.2) and from WhatsApp/Telegram — now applied to the Email channel. It isn’t a text the tenant can configure or turn off.