Roblox Introduces AI Chat Rephrasing to Replace Banned Words
Roblox has launched a new AI-powered chat rephrasing tool that replaces banned words with appropriate alternatives in real time. The feature aims to preserve conversation flow and enhance safety for users, particularly younger audiences. This update follows increasing regulatory pressure and recent lawsuits over online safety for minors.
Roblox has rolled out a new artificial intelligence feature designed to foster a more civil environment in its online gaming platform. The company announced the deployment of a real-time AI chat rephrasing system, which automatically substitutes banned or inappropriate words with respectful and appropriate language during user conversations.
Traditionally, Roblox’s chat moderation relied on filters that replaced restricted terms with hash symbols, which often disrupted the flow of communication and gameplay. The new system, however, enables messages like “Hurry TF up!” to be transformed into a more acceptable “Hurry up!” while maintaining the original intent and clarity. When a message is rephrased, all chat participants are notified, supporting transparency and civility within the community.
According to Roblox, the upgraded AI not only supports all languages available through its own automatic translation tools but also detects attempts to evade filters—including leetspeak, a form of text using letters and symbols to disguise prohibited words. The company reports that the latest system has significantly improved the detection of inappropriate messages, reducing false negatives related to sharing or soliciting personal information by a factor of 20.
Rajiv Bhatia, Roblox's Vice President of User and Discovery Product, stated that the AI-driven chat rephrasing helps maintain both the pace of conversations and community standards. “Real-time rephrasing helps keep gameplay and conversations on track while guiding language toward what's appropriate,” Bhatia said. He highlighted that this approach reduces chat friction and upholds Roblox's commitment to safety.
The new moderation measure arrives amid heightened scrutiny of digital platforms. Recently, Roblox introduced mandatory facial verification for chat access, a move that followed lawsuits filed by attorneys general in Texas, Kentucky, and Louisiana. The legal actions were prompted by concerns about young users facing potential risks such as grooming and exposure to explicit content on the platform.
Roblox’s application of artificial intelligence to language moderation reflects a broader trend within the online gaming sector—using machine learning models to address complex ethical and safety challenges as large, real-time communities interact globally.
Related Posts
Publishers Gain Right to Opt Out of AI Search Following New Regulation
New regulation grants publishers the ability to opt out of having their content indexed or used by AI-powered search engines. This policy shift is expected to reshape the relationship between content creators and major AI platforms. Industry observers note the potential for significant impact on access to information and copyright enforcement.
Amazon’s Ring Faces Lawsuit Over Facial Recognition Data Collection
A class action lawsuit has been filed against Amazon, alleging that its Ring security cameras unlawfully collected and used facial recognition data from millions of Americans. The suit seeks financial damages related to Ring’s Familiar Faces feature, which identifies individuals using AI-based biometrics.
Mathematicians Raise Concerns Over AI’s Impact on Mathematical Research
A group of mathematicians has issued a declaration highlighting threats posed by artificial intelligence to the integrity and future of mathematical research. The Leiden Declaration, developed over several months and endorsed by the International Mathematical Union, expresses concerns about increasing industry influence and recent AI-driven advances such as disproving longstanding conjectures.