Building on the AI Categorization Beta it would be very interesting to see Fin AI involved in spam detection as it seems in some cases very obvious spam makes it through into the Intercom inbox of teams we are working with and also sometimes messages that are from clients also make it into the spam folder.
Even worse if Fin doesn’t recognize it as spam and starts communicating with the spammer and then you are paying for that engagement.
I was wondering how Fin might help with this process:
- Detect it may be spam and indicate it with summary of why it thinks it might be spam
- Not engage with the potential spammer until confirmed that it’s not spam
- Marks it as spam and updates rules to block similar future spam
- Learns from similar activities across thousands of Intercom inboxes to improve on it’s ability to detect spam
Along with this it would be great to have a way to clean the spam inbox when needed, so after checking it for false/positives you can clear the box to prevent having to review the same messages again. Being able to force empty the spam folder it will help a lot with planning a schedule to process and clear.