Vibebo AI Sentinel is Vibebo’s AI-assisted moderation layer for anonymous messages. It classifies incoming text to help detect spam or promotional messages, harmful content and sexually suggestive content before those messages become an unwanted part of your inbox experience.
What Sentinel is for
Sentinel looks at the text and context of a message and classifies it against the filtering categories you choose. This gives you broader protection than a fixed word list while leaving you in control of which categories are enabled.
Test AI moderation with Sentinel
Open the AI Sentinel demo when you want to test sample text and see how the moderation experience classifies it. The demo gives you a practical way to explore AI moderation before deciding which filters you want to use on your own messages.
Configure Sentinel in Message Filtering
Your own filtering controls are in Settings > Safety > Message Filtering. Turn the available categories on or off to match the type of filtering you want for messages sent to your profile. You can change those choices later as your audience or boundaries change.
Sentinel does not identify anonymous senders
Sentinel evaluates the content of a message; it does not reveal who sent it. You can use moderation and safety controls while the sender remains anonymous to you.
Use Sentinel with Hidden Words
Sentinel handles broader content categories; Hidden Words handles the specific words, phrases or emoji you personally choose to filter. Use them together when you want both AI-assisted classification and your own custom boundaries. If an unwanted message still reaches you, Inbox+ also lets you block the sender or report the content.
When to use AI Sentinel
Sentinel is useful when you want broader message filtering than a personal list of blocked terms can provide. It can help when a public link starts attracting repetitive promotional messages, when harmful or abusive language becomes a problem, or when you want to reduce sexually suggestive messages without trying to predict every possible phrase people might use.
How to configure AI Sentinel for your account
1. Sign in to Vibebo and open Settings.
2. Choose Safety.
3. Open Message Filtering.
4. Review the AI-assisted filtering categories available to you.
5. Turn on the categories you want Vibebo to use for incoming messages.
6. Return to the same settings whenever you want to change the level of filtering.
You do not need to enable every category. The right setup depends on why you are sharing your Vibebo link and the kind of audience you are inviting. A creator sharing publicly may want stronger filtering than someone collecting feedback from a small private group.
What happens when you use Hidden Words as well
Hidden Words gives you a more personal layer of control. For example, Sentinel may help recognise a sexually suggestive message even when it uses wording you have never seen before. Hidden Words is better when there is a specific nickname, phrase, slur, topic or emoji that you already know you do not want in your inbox.
The two controls can be used together. Message Filtering handles broader categories; Hidden Words handles the specific terms you choose.
How to use the Sentinel demo
The Sentinel page lets you test text against the moderation experience before changing your own settings. Enter sample wording and review how it is classified. Testing can help you understand the kinds of distinctions AI moderation is trying to make, especially where the same word can be harmless in one sentence and inappropriate in another.
The demo does not change your account settings. To control filtering on messages sent to your own profile, use Settings > Safety > Message Filtering.
If a message still reaches your inbox
No automated filter will remove the need for user judgment in every situation. If a message reaches Inbox+ and you do not want further interaction from that sender, use the block action available with the message. If the content should be reviewed by Vibebo, use the reporting option. You can also update Hidden Words when a new recurring phrase becomes something you no longer want to receive.
Sentinel is designed to make message moderation more manageable while keeping the sender anonymous to you. It evaluates message content for moderation purposes; it does not turn filtering into a way to discover who sent a message.