Roblox launches AI-powered chat system that automatically rephrases inappropriate messages
- At launch, the system focuses mainly on automatically rewriting profanity in chat.
- Instead of showing “#####”, messages such as “Hurry tf up!” may appear as “Hurry up!”.
- Roblox says the system is designed to guide user behaviour rather than simply block messages.
Roblox has unveiled a new AI-powered chat feature designed to automatically rephrase messages containing profanity in real time.
The company said the update builds on Roblox's existing moderation tools as it looks to improve communication and reduce disruptive text filtering on the platform.
Now, instead of blocking messages entirely, the new system rewrites filtered text into more respectful language while preserving the original meaning.
At launch, Roblox said the feature focuses primarily on profanity. For example, a message such as “Hurry tf up!” may be automatically rephrased as “Hurry up!” allowing conversations to continue without the confusion often caused by fully blocked text.
Early rollout
Moreover, the company said early tests show the system improves moderation accuracy, reducing missed attempts to share personal information such as phone numbers or social media handles by 20 times.
“Think of this system as a steering wheel, rather than a stop sign: If a message violates our policy, everyone in the chat is notified that the text has been rephrased to keep things civil," said Roblox in a post.
“By matching restricted words to phrases that users actually use in everyday chat, we help keep chat respectful without losing the authentic feel or flow."
It added: “We’re still at the beginning with rephrasing and providing more context to users when text is blocked. Over time, like any AI translation tool, it will become increasingly accurate and nuanced in how it preserves a user’s original tone or context."