Roblox is implementing the system to discover the children’s warranty chat messages

Roblox is a popular online game platform for children. But it is also a place where people who want to use children know that they can find an audience. On Thursday, Robox announced Roblox SentinelAnd Artificial Intelligence System It is designed for the brand on the inappropriate messages in its chat feature. Roblox is already prohibited to share realistic images and personal information, such as phone numbers and user names. The company hopes that Sentinel will inform more messages sooner than the investigation. It has been played on the platform since late 2024, but it was just announced this week.

A representative of Roblox did not immediately respond to a request for comment.

How Roblox Sentinel works

Roblox Roblox Sentinel, an Amnesty International system to help discover signs of children at risk. Once Roblox realizes the problems, actors can investigate and reports to the law.

SENTINEL manages an actual time analysis across more than six billion chat messages every day. It takes constantly one minute shots, which are automatically analyzed by artificial intelligence to determine messages that may be harmful to children. These messages are also assembled over time to show and report that can be investigated.

SENTINEL flags on their training. He was trained to distinguish between safe messages and those that were previously reported because it violated the policy of children’s enforcement in Robox.

Roblox Sentinel is available as an open source

According to Roblox, so far in 2025, Sentinel has helped discover 1,200 reports on the exploitation of potential children reported to the National Center for missing and exploited children. Roblox hopes to provide Roblox Sentinel like Open source Other companies have to integrate into their systems. The symbol is available on Roblox now.

Roblox is not the first attempt to use artificial intelligence to monitor content and improve online safety. In early July, I shared how it is Using artificial intelligence to moderate content Via 25 languages in real time. AGE verification It is now available for teenagers who want to chat.