JarvisTheChatbot

Content Filtering Update

Hey all, I realise that there has not been any visible updates for a couple of weeks now. Please be assured that this doesnt mean there hasnt been any work going on, myself and aqua have been busy working behind the scenes on some pretty comprehensive systems.

Firstly, we now have the most advanced content filtering system we ever have, detecting everything from possible spam (on 10 different flags) to profanity to low user content, as well as reports you guys give us through the ❗ emote. Coupled with prioritised and automated reviewal, our Jarvion team is able to remove the most explicit and disruptive patterns easier than ever. The system, like Jarvis itself, is constantly advancing based on the approved and denied reports from our Jarvion team, meaning it is constantly detecting more and handling more autonomously on every report submitted.

Secondly, we have a brand new and more balanced blacklisting system with built-in warnings and several temporary punishments before a permanent blacklist from contributing data to Jarvis. Ive also introduced silent blacklists to deal with the handful of users attempting to exploit Jarvis, so they can continue to do so without actually affecting anything (lol)

As these systems are finishing with their beta test and are now on the live bot, we will begin scanning the databank on a massive scale constantly in the background to ensure the latest methods of detection are constantly removing bad patterns from Jarvis, preventing issues before they happen. If you have any questions about these new systems, please tag me in #support and Ill try to answer them to the best of my ability, but I will not be revealing *too many* details as want them to be as secure as possible, thanks all!

Darth | 2 months ago

Back 10