Configurations

Assisted FUD Control

FUD (Fear, Uncertainty, and Doubt) can encompass fear-inducing queries and direct accusations that can undermine a project's credibility. To combat this, we've developed a model capable of identifying and flagging instances of negativity or fear in questions. However, as an AI model, Hawky may occasionally produce false positives. To address this, we've implemented a unique solution involving community assistance to reduce false positives.

How It Operates

When Hawky flags an individual for spreading FUD, a timer starts, signaling that the person will be muted due to their negative message. The community has 15 seconds to intervene and prevent this action. If the community agrees with the flagging, they can take no action, and the member will be muted automatically. However, if the community believes it's a false positive, they have the opportunity to prevent the mute.

Spam Guard

To combat the persistent issue of spam within Telegram communities, we've developed a configurable model capable of handling spam such as excessive promotions and scam links. Spammers are promptly banned to maintain community integrity.